Talk:Steam (service)/Archive 3

Archive 1

Archive 2

Archive 3

Archive 4

Archive 5

→

Archive 9

Bias

the article seems to put a lot more thought and weight into explaining where this system falls short. this was brought to my attention during a philosophy class, from someone who is not much more than an occasional user of wikipedia. after reading the article, i'd ahve to agree with him saying that much of the page is complaining about the weaknesses (and even the strengths) of Steam, rather than just explaining what it is and how it functions. wikipedia is supposed to first answer the question "what", and then, if interest is large enough, answer the question "what does this mean in practice". if i didn't know about steam myself, i would be frustrated because this page does not explain the architecture of the actual system itself.... i apologize if i may be seen as rambling... -- Bob52287 22:20, 23 April 2007 (UTC)

I disagree. The portion of the article above Criticisms is fairly unbiased. The Criticisms portion just takes up a large part of the article. I haven't read that part yet, but if they are valid criticisms, I don't see why they shouldn't be posted in an appropriate way. It seems that they are rather accurate (I don't know about fair yet). Lasdlt 02:54, 24 April 2007 (UTC)

it's not that it shouldn't have criticisms, but the article should focus more about what Steam is, before even mentioning any sort of criticism. a friend of mine didn't know anything about Steam aside from what it looks like and that he has to use it to play Valve games, and when he went to this page, he came away from it still not being sure of what Steam is. in a couple weeks, if nobody's done it yet, i'll work on the article a bunch, give it a good once-over. -- Bob52287 18:40, 27 April 2007 (UTC)

That's an excellent point. I'll see what I can do. :-) --Tom Edwards 12:05, 24 April 2007 (UTC)

I suggest possibly removing the "Recent and notable centralized issues" section altogether in a means of slightly leveling the balance? They're all in the past and don't deal with what Steam actually is. Maybe to keep users informed of current issues, this be updated accordingly, then removed after it no longer matters? Scriptdaemon 19:13, 4 May 2007 (UTC)

That's what's been happening. It may well be time to clean it out again. --Tom Edwards 22:27, 4 May 2007 (UTC)

Well, for an unbiased view on something all the facts and issues must be explored. And STEAM has a lot of criticism thrown it's way, so the long list of it's faults shouldn't be removed.

--joshsaunders 13:05 3 November 2007: this article screams biased, articles should not be written from the point of view of any one demographic and the only demographic that has the problems mentioned in this article are "harcore" gamers, normal casual gamers or non "hardcore" gamers dont see any problems with the system. —Preceding comment was added at 03:09, 3 November 2007 (UTC)

Linux Support

Feel free to added this back if its properly cited, but the "criticism" is based on a bunch of forum posts. There has to be more proof of this said criticism from reliable sources. Also the "banned" comment is from a valve employee and may or may not be the official position of valve. Corpx 05:19, 20 July 2007 (UTC)

Hello Corpx, the fact that the sources come from Value's forum does not negate that they are "primary sources" as per Wikipedia's official policy as described in WP:RS. Masterhomer 05:28, 20 July 2007 (UTC)

• WP:OR says "Primary sources that have been published by a reliable source may be used in Wikipedia" - I removed them due to the lack of secondary coverage Corpx 05:33, 20 July 2007 (UTC)

The secondary coverage is indeed there, one being directly from Value themselves, and one from Digg. I added an additional primary reference directly from Steam's website. Masterhomer 05:35, 20 July 2007 (UTC)

• Secondary reference has to come from independent sources - Digg.com is not a reliable source. Corpx 05:37, 20 July 2007 (UTC)
• The criticism part is purely original research. A bunch of users posting on a company's forum is not criticism. Corpx 05:39, 20 July 2007 (UTC)

You must have a different definition of criticism then what is the norm. Can you elaborate on this? Masterhomer 05:43, 20 July 2007 (UTC)

• "Any interpretation of primary source material requires a secondary source." from WP:OR Corpx 05:48, 20 July 2007 (UTC)

• Per WP:OR: A primary source is valid when: (1) only make descriptive claims, the accuracy of which is easily verifiable by any reasonable, educated person without specialist knowledge, and (2) make no analytic, synthetic, interpretive, explanatory, or evaluative claims.

Also said section includes three primary sources and one secondary source. This makes it very well referenced, especially when compared to the reference density of the rest of the article. The combination of these sources makes it trivial to verify the facts as described in the section. Therefore it should comply with Wikipedia's polices. Also remember WP:GF. Masterhomer 05:57, 20 July 2007 (UTC)

• Digg.com is not a secondary source. Corpx 05:58, 20 July 2007 (UTC)

I've fixed up some factual inaccuracies. Whoever wrote the section needs to calm down a little and do some proper research. --Tom Edwards 16:35, 20 July 2007 (UTC)

• Again, this is a primary source and we're interpreting the criticism from it. Digg.com is not a reliable source, so I dont even see why its mentioned here. A bunch of people posting on a forum does not equal criticism. A media source has to write about it for it to be criticism Corpx 16:51, 20 July 2007 (UTC)

When it comes to criticism it doesn't matter so much if it's a factually reliably source (so long as the article itself gets the facts right). What matters is that there are significant numbers of people criticising. In my mind, there have been more than enough. (BTW, could you stop using bullet points please? It makes things harder to read.) --Tom Edwards 07:33, 21 July 2007 (UTC)

I dont "user-edit sources" are reliable sources at all. If there was enough criticism, it would've been picked up by a reliable source. There are lots of user complaints against steam, but I dont think we should list them all Corpx 16:29, 21 July 2007 (UTC)

Reduced Performance

When Valve moved GoldSrc to Steam they also made a raft of other upgrades and changes. Steam itself does not affect loading times in any appreciable way (unless you count the time it takes to load itself), as can be seen by comparing any other game that has existed apart from Steam, or, dare I say it, been cracked.

I'm not happy with this edit, either. It's saying the same thing as before, only with twice as many words! Fixed. --Tom Edwards 18:55, 21 October 2007 (UTC)

Your dispute of this revised text is unjustified. You are also a Valve shill

I don't find that a terrifically convincing argument. --Tom Edwards 19:19, 21 October 2007 (UTC)

It is not surprising that a Valve shill like you would find it unconvincing. However the fact is that the system requirements for Valve's games released prior to Steam are significantly less demanding than the versions of the same games which require Steam, and Steam's overhead is responsible for that. Even your obvious bias for Valve can't escape the basic fact that Steam is one additional application which must be running on a PC, and it consumes precious computer resources (RAM, CPU cycles, network processing, etc.) that would otherwise be available to the game itself. For example, that is why a PC which meets the minimum specifications to run the original released version of Half-Life 1 will not run the newer version which requires Steam.

Steam does reduce the performance of games, and it can be seen by comparing the versions of games which don't require Steam with those that do. Sure the amount of reduced performance varies by game and the PC tested, but it is still present. —Preceding unsigned comment added by 70.232.52.109 (talk) 01:45, 22 October 2007 (UTC)

Sure, it takes up a few more cycles. But there's a point at which that becomes a noticeable problem worth mentioning on Wikipedia and Steam is a very long way from it. I again ask you not to compare Valve's Half-Life games, which were modified in lots of other ways too, but instead games that have been ported to Steam otherwise untouched, or perhaps games (HL1 if you like) that have been cracked to have Steam forcibly removed. --Tom Edwards 09:13, 22 October 2007 (UTC)

>"Sure, it takes up a few more cycles."

In the case of some games and some computers it is more than just a "few more" cpu cycles. It is more CPU, memory, and network useage all of which affect performance. However, the objectively minded will notice how your bias for Valve and Steam causes you to misrepresent and trivialize the real negative effects I mentioned.

>But there's a point at which that becomes a noticeable problem worth mentioning on Wikipedia and Steam is a very long way from it.

That is only in your biased opinion. The fact that I am not the one who originally added this Steam peformance criticism section to Steam's Wikipedia article but merely editing it for clarity and better accuracy clearly demonstrates that I am not alone in the rational assessment that Steam can cause reduced game performance. Therefore that criticism should be included in the Wikipedia entry.

>"I again ask you not to compare Valve's Half-Life games"

There is no legitimate reason to not include them as they clearly demonstrate the fact of reduced performance.

>"which were modified in lots of other ways too"

Aside from being modified to require Steam, the original Half-Life games have NOT been substantially modified otherwise. Only minor bugs and exploits have been fixed, and many if not most of those fixes were Steam-related and therefore not found in the pre-Steam versions anyway. Therefore it is certainly fair to compare the Steam versions to the releases which preceded Steam.

>"but instead games that have been ported to Steam otherwise untouched"

Even if those games showed no performance decreases (and I doubt that is the case for every one of those games), that does not negate the fact that Valve's original Half-Life games do have such performances decreases. You can't exclude some games from consideration because they don't support your biased position.

>"or perhaps games (HL1 if you like) that have been cracked to have Steam forcibly removed."

Comparisons should only be done with official releases not so-called "cracked" versions since such tampering and changes to the games by third-parties are unauthorized and would most likely skew the results. The only party which can unequivocally remove Steam from a game is Valve. —Preceding unsigned comment added by 70.232.52.109 (talk) 11:16, 22 October 2007 (UTC)

Even if those games showed no performance decreases (and I doubt that is the case for every one of those games)

I'm afraid you will find that it is the case, for all of them. The same binaries are loaded in the same way for every game that integrates, GoldSrc included.

No, it's NOT the case. Executables ("binaries") for games integrated with Steam are both linked to the Steam.dll file and are protected with PE Compact to prevent reverse engineering. They are NOT the same binaries are those of non-Steam releases of the same games as those binaries do not have the overhead of Steam.dll or the rest of the Steam software.

Aside from being modified to require Steam, the original Half-Life games have NOT been substantially modified otherwise.

Do you have any proof of that?

>"Could you perhaps explain to me why GoldSrc is an exception?...Do you have any proof of that?"

Yes. First, look at the change logs for updates to the Half-life 1 games. The changes implemented for the Steam versions have been minor and pertain to security fixes and Steam-specific additions and corrections. Second, the media files (sound, maps, models, textures, etc.) in the Steam releases of the game are identical to the non-Steam releases. The only substrantial differences between the two versions of each game are Steam-related. Therefore, the additional overhead of Steam is responsible for the performance issues and additional system requirements.

The fact that I am not the one who originally added this Steam peformance criticism -section to Steam's Wikipedia article but merely editing it for clarity and better accuracy clearly demonstrates that I am not alone in the rational assessment that Steam can cause reduced game performance.

You and another anonymous IP are the only ones claiming reduced performance. I would like to see some recent third-party complaints that can be used as sources.

Although I have done no third-party game testing, third-party complaints are not necessary to establish reduced performance as a legitimate complaint. If Steam reduces the performance of ANY game, the complaint is genuine and should be included. In addition even logic should tell you that running an additional application like Steam while running a game will cause at least some level of performance degradation versus running that game alone as the Steam application uses CPU, RAM, and network resources while it is running even while minimized.

FYI, the majority of the Criticism section is my writing. It would also be helpful if you would stop personalising this discussion. --Tom Edwards 14:02, 22 October 2007 (UTC)

The majority is your writing because you want to minimize and downplay the complaints. I am rightfully pointing out your bias because it is obvious through your actions and long history of editing the Steam and Valve-related Wikipedia entries.

If you have no further arguments, I'll revert the changes later today. Missed your answers again. Please see my Survey results response to help prevent this in future. --Tom Edwards 15:28, 23 October 2007 (UTC)

The changes implemented for the Steam versions have been minor and pertain to security fixes and Steam-specific additions and corrections.

The major changes were made when the move first took place and are not included in any changelog. Look for yourself and you will see no patch notes for the public release of GoldSrc on Steam, only the subsequent changes made to it. You will also find in those changelogs that Valve introduced performance fixes in GoldSrc itself - something that would not be possible given your assertion that Steam is the only thing slowing the code down, as it is updated separately.

You will also find in those changelogs that Valve introduced performance fixes in GoldSrc itself - something that would not be possible given your assertion that Steam is the only thing slowing the code down, as it is updated separately.

There you go again. My assertion is NOT that Steam is the ONLY thing slowing the code down. It is simply a main thing because running Steam consumes resources (CPU, RAM, Network). The Steam.dll is linked into the Steam version of the Half-Life 1 engine when it is compiled. So, Steam is run when the game is run. Certainly Valve could have made some minor performance improvements to the rest of the Half-Life 1 code, but those improvements don't preclude Steam's performance impact. If the Steam.dll and the rest of Steam were removed from the game, the game's performance would be greater and any other performance improvements would have more impact.

Executables ("binaries") for games integrated with Steam are both linked to the Steam.dll file and are protected with PE Compact to prevent reverse engineering.

Does the PE Compact code change between games?

If Steam reduces the performance of ANY game, the complaint is genuine and should be included.

I assert that this is not true for your example of GoldSrc on two grounds:

1. GoldSrc games are the only ones affected
2. GoldSrc games were modified in other ways when transitioned to Steam

They're not. Steam adds more overhead to the game because of all of the extra Steam code (Steam.dll etc) that is running while the core game is running. But, the media assets which are the main determinant of the game's performance are no different. None of the game's core functions such as the renderer and audio system has been changed. To put it in simple terms, Steam reduces the performance of the game because it steals resources from the game. That is why a PC which meets the minimum specifications of the non-Steam version of Half-Life 1 won't run the Steam version as well if at all.

Your best bet to respond to this argument is finding non-GoldSrc games with reduced performance under Steam. If your argument is correct, this should be easy enough. --Tom Edwards 15:58, 23 October 2007 (UTC)

My assertion is NOT that Steam is the ONLY thing slowing the code down ... Steam reduces the performance of the game because it steals resources from the game.

Why does Steam take more resources from GoldSrc than other games? Or is that not your claim any more? Why not answer all of my questions at once by producing citable evidence of these effects in other games? --Tom Edwards 13:39, 26 October 2007 (UTC)

Why does Steam take more resources from GoldSrc than other games? Or is that not your claim any more?

That was never my claim. My point was and is that Steam negatively impacts the performance of Half-Life 1 games, and this impact is more noticeable than with most if not all newer games because the system requirements for the original releases of the Half-Life 1 games are so relatively low. Because Steam consumes resources and it is running when all Steam games are run, it does negatively impact the performance of ALL Steam games to some degree. However, that impact can be negligible depending on the game and the PC on which the game is being run. My point with citing the Half-Life 1 games in the "Reduced Performance" description is to give an example where that performance reduction is most pronounced. (Now let's see if you will finally read and understand this point which you have forced me to needlessly repeat multiple times on this page.) —Preceding unsigned comment added by 70.232.93.130 (talk) 06:50, 27 October 2007 (UTC)

None of the game's core functions such as the renderer and audio system has been changed.

Can you prove that? --Tom Edwards 13:39, 26 October 2007 (UTC)

However, that impact can be negligible depending on the game and the PC on which the game is being run.

You now appear to admit that Steam's effect is only notable on old computers and/or when playing old games. Is this not simply another way of saying that it increases minimum requirements? --Tom Edwards 10:44, 27 October 2007 (UTC)

First, I am not claiming anything new. You just simply didn't read and understand what I wrote before due to your bias for Steam. Second, yes, Steam does increase the minimum system requirements BECAUSE it consumes resources not simply because Steam itself has higher minimum system requirements than the non-Steam Half-Life 1 games.

Note that GoldSrc is still an anomaly. You will not find the same order of performance drop from other games regardless of their age. --Tom Edwards 10:52, 27 October 2007 (UTC)

Oh yes Steam does have the same effect on other such games. Sin 1 is one, and the Steam version of the game is basically identical to the updated non-Steam version except for the linking to the Steam.dll/the requirement that Steam be running. However the number of such games is limited because few if any additional games on Steam are of the same age as the Half-Life 1 games.

Excellent, you've given a second source. I've never played Sin 1; do you have a citable source for its performance loss? How did you hear about it?

Steam does increase the minimum system requirements BECAUSE it consumes resources not simply because Steam itself has higher minimum system requirements

If Steam takes resources from the system during play, then a surely performance drop should be noticeable on all games running on older systems? And surely demanding games running on modern systems should show a drop too?

Both of those suppositions are certainly possible, but as I noted above and in my edit the extent of the impact would depend upon the game and the system and their configurations, such as how "old" is older, and what video settings are used in the games. However I am not aware of any extensive testing of all or most Steam games which has been done on this issue or at least of any publicly published results of such testing. That is why I am content to leave the wording to more ambiguous terms, i.e. "can negatively impact a game's performance" not "does negatively impact a game's performance." -70.232.93.130 23:42, 27 October 2007 (UTC)

Lastly, stop removing the disputed tag from the section. I shouldn't have to explain why. --Tom Edwards 20:42, 27 October 2007 (UTC)

I shouldn't have to tell you why the disputed tag does not belong there, but I will. It's because you are the only one disputing it, and your basis for doing so is biased and not rational. Steam does negatively impact the performance of some games on some computers especially Valve's Half-Life 1 games on PC's which barely meet the games' minimum specifications. I am not claiming that Steam impacts the performance of every game on every PC, and it is certainly not necessary that Steam negatively impact the performance of every game for its impact on any game to be a valid criticism worthy of note in the article. In addition even the arbitration panel (which I was unaware of before now) ruled that the "dispute" label on the section was premature, i.e. "* Decline. Premature (as well as content dispute.)" - 70.232.93.130 23:43, 27 October 2007 (UTC)

I shouldn't have to tell you why the disputed tag does not belong there, but I will. It's because you are the only one disputing it, and your basis for doing so is biased and not rational.

You have no authority to make that decision, and it is extremely arrogant of you to do so. There is a dispute ongoing; the text is therefore disputed.

I am not claiming that Steam impacts the performance of every game on every PC, and it is certainly not necessary that Steam negatively impact the performance of every game for its impact on any game to be a valid criticism worthy of note in the article.

Might I suggest moving it underneath the "Changes to minimum specification" heading then?

In addition even the arbitration panel (which I was unaware of before now) ruled that the "dispute" label on the section was premature, i.e. "* Decline. Premature (as well as content dispute.)"

They are not referring to the addition of the tag but the RFA 'dispute' itself. --Tom Edwards 18:06, 28 October 2007 (UTC)

Seconded. Until this issue is resolved, please do not remove the disputed tag. — HelloAnnyong [ t · c ] 18:08, 28 October 2007 (UTC)

Both of those suppositions are certainly possible, but as I noted above and in my edit the extent of the impact would depend upon the game and the system and their configurations, such as how "old" is older, and what video settings are used in the games. However I am not aware of any extensive testing of all or most Steam games which has been done on this issue or at least of any publicly published results of such testing. That is why I am content to leave the wording to more ambiguous terms, i.e. "can negatively impact a game's performance" not "does negatively impact a game's performance."

Thank you for being civil. This is a rational suggestion that I support, though I would add to your suggested text the qualifier of "on older systems through its requirement to remain running as a background process" that you have agreed at different points above exists - we are allowed to be ambiguous in the absence of evidence, so long as we know for certain that what we are being ambiguous about is true within the meaning of our text. Leaving out these qualifiers creates a misleading impression. --Tom Edwards 18:14, 28 October 2007 (UTC)

Fourth opinion

I've yet to see anyone bring up some valid, verifiable references to back up their arguments here. No one has produced hard numbers that proves that Steam reduces performance, nor are there any citations for the claim that the legacy games haven't been modified except for a requiremnt for Steam. Until someone can produce those links, the "Reduced performance" section should be kept off the article (pending the removal of the page protection, of course). That's what the Talk page is for: to hash controversial additions out without getting into a revert war. — EagleOne\^Talk 17:13, 29 October 2007 (UTC)

Seeing as how some of these games are not designed to operate on newer hardware and Windows XP, I still don't understand after reading the huge argument above, how exactly a negative impact may be as bad as it is said. Obviously with Steam or any other program (Winamp, Drivers, Utilities, Windows Services) more CPU cycles and resources are used, but at the same time it is available in the first place. There is a relative limitation to support for lower-end machines, especially with older Operating Systems (See Changes to Minimum Requirements). While I agree that it does reduce performance on extremely lower-end machines due to logical understandings of hardware and whatnot, I do believe that the listed requirements on Steam's Site for an older game (Quake for example), it is taken into account that any machine capable of running the required Operating System (Windows 2000, XP, or Vista) is quite capable of running the older game, while also having sufficient overhead to coupe with the Steam resource use.

On top of my observations, I also notice that the way the portion is written so that all but one sentence is used to describe the negative impacts of having an additional piece of software running. I feel that for this portion to be more fair, it would need to be something more towards understanding that it is an additional program running in the background.

— Freyar 04:58, 30 October 2007 (UTC)

While I agree that it does reduce performance on extremely lower-end machines due to logical understandings of hardware and whatnot, I do believe that the listed requirements on Steam's Site for an older game (Quake for example), it is taken into account that any machine capable of running the required Operating System (Windows 2000, XP, or Vista) is quite capable of running the older game, while also having sufficient overhead to coupe with the Steam resource use.

Actually no, it isn't taken into account, and that is part of the problem. For example, many of the third-party games on Steam such as Psychonauts ( http://steampowered.com/v/index.php?area=game&AppId=3830&cc=US ) especially the older games would otherwise run under Windows 98 and even list it as a supported OS even though the Steam version of the game won't run on Windows 98 because Steam itself won't run under it. These games simply have the system requirements for the non-Steam releases pasted onto the Steam version information pages. -70.232.56.5 08:19, 30 October 2007 (UTC)

I would say that's a problem with the website, not Steam itself. --Tom Edwards 12:55, 30 October 2007 (UTC)

And besides, you wouldn't actually be able to buy the game because Steam would error out before it even started loading. --Tom Edwards 12:57, 30 October 2007 (UTC)

I also notice that the way the portion is written so that all but one sentence is used to describe the negative impacts of having an additional piece of software running.

First, it is in the criticism section of the article. Second, the current wording already does make it known that Steam is running along with the game. However, to say that Steam is only running in the background would be misleading since the main core Steam.dll is actually linked to the game's executable just like other libraries such as DirectX so a main part of Steam is also running in the foreground with the game. -70.232.56.5 08:17, 30 October 2007 (UTC)

I think he meant "in the background" as in "not what the user is specifically interacting with". The Community overlay confuses that somewhat, but then again it is an optional component. --Tom Edwards 12:33, 30 October 2007 (UTC)

This is a bigger problem than it might appear because of Valve's transition from WON to Steam for online play of their HL1-era games (which are still the most popular in the world behind MMOs). They started to require everyone to move from the old version of the games to the new Steam version, and for whatever reasons (Steam plus other engine changes, as I think 70 eventually admitted) performance was noticeably worsened. This all happened a long time ago, of course... --Tom Edwards 12:52, 30 October 2007 (UTC)

Survey results

Undid because it is not definite since the hardware survey results which preceded the minimum specification changes are not available. Current survey results do not truely apply.

The article used in reference was written before the spec changes took place, and the figures included in it are from that time. --Tom Edwards 20:54, 21 October 2007 (UTC)

That is wrong. The article used in reference contains absolutely no facts or statistics to substantiate that the number of Steam users impacted by the changes in minimum specifications was "definitely" small. The referenced article is simply an announcement of the drop of Windows 9X and ME support and does not contain justification for it. The only Steam hardware survey results publically available are the ones I linked, and because they do not contain statistics on the number of Windows 9X and ME users since Steam no longer runs on those OS versions, the "most likely affected" phrase I used to describe the numbers of users affected by that change is more accurate given the lack of definitive numbers available.

Are you reading the same article I am?

No, I was reading the news of the notice on Valve's official Steam website not that third-party site. Valve's site has no such statistics supporting the news. I am not familiar with steamreview.org so I can't judge whether it is a reliable source or not, but I suppose the numbers it reports for Windows 98SE users are accurate from the Steam hardware survey at that time. There are no results there for Windows 95 and ME though.

>"That's a good point, though statistically the chances of thousands upon thousands of Win98/ME users choosing not to submit the survey while their Win2000+ friends did is pretty damn close to zero."

There you go again jumping to unsupported conclusions. There are no statistics reported in the Steam hardware survey which even suggest let alone substantiate what percentage of Steam users actually submit to the survey. There is also certainly no evidence to support your speculation that users of the various versions of Windows submit to the survey in equal numbers. Therefore no one really knows with certainty how many or what percentage users were affected by the termination of Windows 9X and ME support for Steam. That is exactly why I wrote "most likely."

>"I'd settle for "extremely likely".

"Extremely" is too strong, and it is not grammatically correct within the sentence, i.e. "Although both of these new criteria most likely affected only a small percentage of Steam users...". "most likely" fits best for accuracy and grammar.

There are no statistics reported in the Steam hardware survey which even suggest let alone substantiate what percentage of Steam users actually submit to the survey. There is also certainly no evidence to support your speculation that users of the various versions of Windows submit to the survey in equal numbers.

There is no reason why a user of one operating system would be less likely to submit to the survey than a user of another. The same goes for graphics card and numbers of speakers - they don't affect the likelihood of someone responding. If this weren't true, random sample surveys would not be acceptable forms of research.

You are missing the point. There is simply no evidence or statistics in the survery to support your speculation. Surveys are only reliable when they are truely representative of the whole. The Steam Hardware Survey is not a scientific one. There is nothing about the Steam Hardware Survey that makes it truely representative. It is certainly reasonable to speculate that those users who are more proud of their PC's because they have more up-to-date hardware and software are the ones who respond to the survey as a means of demonstrating that pride, and therefore the survey results are skewed in favor of newer hardware and versions of Windows. In any case there are no firm or definitive statistics that indicate how many Steam users were running Windows 95, 98, and ME prior to the change in Steam's system requirements. Therefore, "most likely" is a better way to phrase that uncertainty.

If you believe this is not the case, that there is something skewing the results and reducing the numbers of Win98/ME respondents, it is up to you to provide evidence of that. Please do so now.

No, it is up to you to substantiate your definitive claim. You have not done so.

"Extremely" is too strong, and it is not grammatically correct within the sentence, i.e. "Although both of these new criteria most likely affected only a small percentage of Steam users...". "most likely" fits best for accuracy and grammar.

Call me an optimist, but I think it's within our means to change a few words around... --Tom Edwards 14:10, 22 October 2007 (UTC)

There is no need to change it. In addition the number of users affected is irrelevant to the real point made by the statement which is that the change in system requirements can prevent Steam users from running their existing games. Therefore, the portion of the sentence regarding the amount of users affected could be eliminated. The only reason to include it would be to show a balanced perspective that the negative impact of the changes is not widespread, but that balance must still be tempered with the fact that the extent of the impact is not truely known since there is no definitive data on it.

I'm not complaining about the content, with the exception of ambiguity about unambiguous data, so much as the style. It's long and confusing; "Steam" is repeated four times in one sentence for a start.--Tom Edwards 15:36, 23 October 2007 (UTC)

It's is not long and confusing. It is thorough and explicit. "Steam" is repeated to leave no uncertainty whereas a pronoun might.

How about you try writing a better version here, on the Talk page, then we can discuss it. Here is my version, for reference. --Tom Edwards 15:30, 23 October 2007 (UTC)

"No, it is up to you to substantiate your definitive claim. You have not done so."

My "definitive claim" is that random-sample surveys are reliable. This is the assumed norm in most modern societies, and the basis of much accepted scientific research.--Tom Edwards 15:36, 23 October 2007 (UTC)

You are WRONG. You need to learn the difference between a scientific survey and a non-scientific one. Purely random surveys like the Steam Hardware Survey are NOT reliable nor accepted for scientific research for the very reason I listed above and which you conveniently overlooked. A survey is only reliable when it is truely representative of the whole. The Steam Hardware Survey is NOT representative; it is not a scientific survey. Read what I wrote above instead of repeatedly spewing your same biased, erroneous bullshit in response.

Your claim is a lower proportion of people using Win98/ME submitted than those using Win2000+. This is not normal and you need to explain why it happened. --Tom Edwards 15:32, 23 October 2007 (UTC)

No, my claim is three-fold. First, the Steam Hardware Survey prior to the drop of Windows 98 and ME support only reported the results of Windows 98SE users. Windows 95 and ME were not counted. Second, Windows 95, 98, and ME users may have responded to the survey in numbers that were disproportionate to those of newer versions of Windows simply because their PC's were old and unimpressive. Third, because the Steam Hardware Survey isn't a scientific one but is instead totally random and voluntary, it isn't necessarily representative of the whole of Steam users. Therefore what all of this proves is that it is impossible to definitively know exactly what percentage of Steam's users were impacted by the drop of Windows 98 and ME support. It is most likely a small percentage, but the exact percentage is not known since the Steam Harware Survey is optional, incomplete (only Windows 98 counted), and not representative.

It is certainly reasonable to speculate that those users who are more proud of their PC's because they have more up-to-date hardware and software are the ones who respond to the survey as a means of demonstrating that pride, and therefore the survey results are skewed in favor of newer hardware and versions of Windows.

Wikipedia does not speculate in any form, so no, this is not reasonable at all. I also find it amusing that you accuse me of unsupported speculation and then present this as an argument. --Tom Edwards 15:36, 23 October 2007 (UTC)

You are obviously an idiot. Learn to read what I wrote in the context in which it is written. The point I am making with that speculation is to refute your erroroneous claim above that "There is no reason why a user of one operating system would be less likely to submit to the survey than a user of another. The same goes for graphics card and numbers of speakers - they don't affect the likelihood of someone responding." The fact is that it is very possible that those users who are more proud of their PC's because they have more up-to-date hardware and software are the ones who respond to the survey as a means of demonstrating that pride, and therefore the survey results are skewed in favor of newer hardware and versions of Windows. So, to put it in simple terms you can understand, some users might not submit to the survey because they are embarassed of their PC's, and that would skew the results in favor of newer PC's. So, there IS a reason why those users would be less likely to submit, and that is by no means the only possible one.

You haven't answered by argument. You need to provide evidence of this, or it is no less accurate than every other random-sample survey, which also face the same issues. --Tom Edwards 13:41, 26 October 2007 (UTC)

I'm reverting to the shorter version, as you've now made multiple edits to the above discussion without saying anything here. --Tom Edwards 20:49, 27 October 2007 (UTC)

I made no further comments here because I wrongfully assumed you had finally come to your senses and had stopped contesting my rewrite on this. Your position on this is simply without merit. I see no reason to keep repeating myself when your bias simply will not allow you to see the validity of my position. The Steam Hardware Survey is NOT a scientific survey. It is incomplete. It is NOT representative. Therefore it is NOT a truely reliable source for a definitive answer to the question of how many Steam users were affected by the changes in Steam's system requirements. Despite your point that similar unscientific surveys are also flawed in the same way doesn't make the use of the Steam Hardware Survey any more reliable or definitive. In addition, we don't even have official results of the Steam Hardware Survery from immediately before each of the changes in system requirements. The third-party website referenced is not an official source, and even its reported results are from March 1, 2006 and not June 30, 2006 when the requirements were actually changed. Neither you nor I nor probably even Valve knows how many or what percentage of Steam users were running Windows 95, 98, and ME when support for those versions was dropped. It was most likely a small percentage, and the wording of my edit explicitly states that uncertainty. 70.232.93.130 22:38, 27 October 2007 (UTC)

The Steam Hardware Survey is NOT a scientific survey. It is incomplete. It is NOT representative.

This is why the shortened version of the text reads "according to". Frankly, I'm tired of trying to argue over semantics and will leave this as it now; it's not exactly giving an incorrect representation in its current state. Perhaps once you've tired of editing the article or are banned someone else will come along and clean it up for us. Thank you for inserting your comment properly! --Tom Edwards 18:00, 28 October 2007 (UTC)

Privacy concerns

This article makes the claim the data collected by Steam is "anonymous" and "non-identifying". Where are the supporting references? Issues like this should be front-and-center things; if existing references used here are sufficient, repeat them, with more detailed location references (page numbers, if not direct quotation). My guess is that the only reliable references are the claims made by Steam itself. If this is in fact the case, then in addition to these references, the article needs to be adjusted to reflect the fact that they are only assertions, not incontrovertible facts. mdf 13:04, 24 October 2007 (UTC)

It's in the SSA and Valve's Privacy policy (I'll rephrase the section to include this now). Valve would be breaking its agreement with users if it violated either, and we'd need references to support that. --Tom Edwards 10:35, 25 October 2007 (UTC)

Turns out there were some empty refs to fill. :-) --Tom Edwards 10:41, 25 October 2007 (UTC)

Thank you for those references! Reading them, though, one immediately encounters this: "By using Valve's online sites and products, users agree that Valve may collect aggregate information, individual information, and personally identifiable information, as defined below." Emphasis added, as it now seems the in-article statement is incomplete. Also, the "personally identifiable information" is said to be "voluntary", but then again, the statement goes on to qualify itself "for certain products and online sites, Valve's collection of personally identifiable information may be a requirement for access to the product or site." And that this "personally identifiable information" can in fact be shared with anyone Valve likes ("Valve may allow third parties performing services under contract with Valve to access stored information but such access shall only be to the extent necessary to provide those services.")

This may be my perspective, but anyone who refuses to provide information and is subsequently rejected for service by Valve, the details of this referenced policy are practically moot. So when writing an article on this subject, it seems to me it should focus on those who are subject to the policy, as the freedom to choose to associate with Valve was not created by this policy.

Interestingly, this policy has no mention of sanctions, liabilities, or other remedies should Valve or any of its sub-contractors violate it. Which brings me to the last point: as I guessed, these are all just promises by Valve. Because of the proprietary nature of the system, nobody except Valve knows what is being sent back and forth, what data is retained, lifespan, etc. No one knows who Valve's sub-contractors are. No one knows if Valve isn't putting all the information on a DVD shipping it to (say) the NSA. While it is true that could be construed as a "violation" of the policy, and it may well be in Valve's interest to abide by the policy (albeit, one they can change at their whim) the fact remains that at the end of the day, that until substantiated by Valve by releasing the internal nitty-gritties of their customer surveillance program for public consumption and analysis they are simply assertions, and I say this section of the article would do better to reflect that reality. mdf 13:18, 25 October 2007 (UTC)

I covered privacy policies in an ICT module a couple of years ago and none of this is remotely unusual. No sane company will restrict itself to a preset list of companies who it sends data to because of the faff of updating that list and waiting thirty days every time it needs to change - it's just assumed that the user will trust you not to throw briefcases of credit card numbers off a cliff (in England, though perhaps not America, there are overriding Data Protection Act laws that force data holders to only allow access to others for legitimate reasons too). Unless there's some specific, citable reason to make a deal of it I don't think it's worth more than a brief sentence. --Tom Edwards 15:47, 25 October 2007 (UTC)

Third opinion

Hey. I just saw that there's an open request for a third opinion for this page. Do you still need help, and if you do, can someone give a decently concise (and NPOV) explanation of what's going on, and what the debate is? Thanks! — HelloAnnyong [ t · c ] 15:01, 27 October 2007 (UTC)

There's a description here. Posting that RFA has coincided with a change of tone from the IP (what good luck!), so things aren't quite as bad as the description makes out right now. --Tom Edwards 16:43, 27 October 2007 (UTC)

Alright. Since there's an RFA out there, I'm going to remove this listing from the 3O page. Thanks for clearing that up. — HelloAnnyong [ t · c ] 16:51, 27 October 2007 (UTC)

Heh. I actually _read_ the RFA, and I see they rejected it. Do you still need help here, or is the situation resolved? — HelloAnnyong [ t · c ] 16:54, 27 October 2007 (UTC)

No, it's still going on above. See #Reduced Performance. --Tom Edwards 20:43, 27 October 2007 (UTC)

Since the RFA is closed, I'm going to take a shot at giving a third opinion here. Let's see.

• 70.232.93.130, you need to rethink your strategy here. First off, you violated Wiki's three-revert rule, which is a bannable offense. Second, you need to start signing your comments on here. I had a really hard time reading the talk page since I can't tell whose comments are whose. Sign your comments by using four tildes (~~~~). Third, stop reverting and actually discuss the issues here.
• This seems to be an issue of WP:OR. The entire reduced performance section is uncited, and I see no reason not to have the disputed tag on there until it gets cleaned up.

• I'm bothered by the tone of the criticism section. Statements such as "The full extent of undisclosed data collection is unknown", "However, this speculative public statement attributed to Mr. Newell is not contained or otherwise reflected in the current Steam Subscriber Agreement (found at http://www.steampowered.com/v/index.php?area=subscriber_agreement)", "must currently be weathered by users wishing to play online or make new purchases. Recent examples of such downtimes are listed in the History section", and the entire reduced performance section are inappropriate and need to be changed.

Admittedly, I'm sort of lost by the constant revert/edits that have been made here. I read the RFA and I've looked through the other sections, but I'm still not really sure exactly what I can do to help. I'm sorry I can't be of more help immediately, but I want to help get this reolved, so please let me know what specifically needs to be commented on. — HelloAnnyong [ t · c ] 21:08, 27 October 2007 (UTC)

That covers the majority of it, thanks. I think it speaks volumes that you include some of my own copy as being of questionable tone - what with me being the biased Valve shill and all. ;-) We'll have to see how our anonymous friend responds... --Tom Edwards 21:30, 27 October 2007 (UTC)

70.232.93.130, you need to rethink your strategy here. First off, you violated Wiki's three-revert rule, which is a bannable offense. - HelloAnnyong

As did you and Tom violate it in reverting my comments. It works both ways, see? Tom's having you revert my edits for him without discussion may technically skirt the rule but not its intent. -70.232.93.130 22:16, 27 October 2007 (UTC)

• I'm bothered by the tone of the criticism section. — HelloAnnyong

LOL! The tone of the criticism section should be critical. It should and does list criticisms reported by Steam users including that of reduced performance. As with any criticism, certainly there is some element of dissent or bias. The problem with at least certain parts of the criticism section right now is that Tom is so biased that he eliminates all reports of criticism that he refuses to acknowledge due to his bias for Valve and Steam, and rewrites the rest of the criticism to gloss over and mitigate its actual extent and impact. Tom hovers over Valve and Steam-related entries in Wikipedia like a mother hen and persistently swats away all critical entries until their authors give up and move on. And, judging from his history of edits on this article alone, he has been doing this religiously for over two years. Tom's bias for Valve and Steam is obvious from his long history of doing this, and it has compromised the accuracy of this article. I' m not looking to rewrite this whole article, but it certainly wouldn't hurt if someone other than Tom or his ilk did. 70.232.93.130 22:16, 27 October 2007 (UTC)

Again, stop putting your comments in the middle of other comments! Please stick to Wikipedia conventions.

No one is arguing against the criticism section, but like everything else on Wiki, it has to maintain standards: that of verifiability, no original research, and appropriately cited. Statements MUST BE CITED. That's not arguable. Furthermore, comments such as "clearly demonstrate" fall within the realm of original research. If you can find a verifiable third-party source that makes claims about Steam, then by all means, add them!

Let's talk about the Changes to minimum specifications section. The first paragraph is fine to me, but the second paragraph states the following:

these changes in system requirements clearly demonstrate how Valve's policy of forcing all Steam users to run only the latest version of the Steam software can prevent existing customers from being able to play their purchased games because their computers can no longer run Steam due to its changed system requirements.

What you're doing here is taking information - namely, that Valve is increasing the minimum requirements, and that these changes affect very few users - and drawing a conclusion. WP:SYN basically covers this, and says that it's inadmissible. — HelloAnnyong [ t · c ] 22:33, 27 October 2007 (UTC)

and drawing a conclusion. — HelloAnnyong

First, I am not the original author of that conclusion. I simply reworded it for better accuracy.

Second, the conclusion is factually accurate, and it does belong because it fits contextually with the rest of the criticism section in which it is located. I agree that the conclusion wouldn't be appropriate were it not for its location. In this case the mere fact that Steam's system requirements were changed is not the criticism. The criticism is how those changes negatively impact Steam users, i.e. WHY the changing of the system requirements is a criticism. And, the answer is what I wrote. It is a criticism because first, it raises the barrier to entry in that only consumers with newer PC's and operating systems can use Steam. Second, it is also a criticism because existing Steam users who do not upgrade their PC's and/or operating systems will be unable to continue to play the games they purchased which require Steam. Both of those are statements of fact, and as these comprise a real criticism of Steam, they deserve mention in this Wikipedia article as such. 70.232.93.130 23:16, 27 October 2007 (UTC)

You're missing the point. In that section, you're taking two sources and drawing a conclusion based on them, which is not allowed as per WP:SYN. Find a credible source that states what's written there, and it can stay. Furthermore, that sentence is worded very strangely/obtusely, so I'm going to attempt to rewrite it. — HelloAnnyong [ t · c ] 17:01, 28 October 2007 (UTC)

You're missing the point that what you are deeming a "conclusion" is simply a statement of two facts which are supported by the preceding and cited statement of fact. The two facts simply explain the cited one. Needing to cite an additional source to support every factual statement is ridiculous. If every factual statement written in a wikipedia article had to first be stated by some outside source, wikipedia articles would be excessively brief and incomplete. -70.232.93.130 21:03, 28 October 2007 (UTC)

Read WP:NOT#OR: "Wikipedia is not a place to publish your own thoughts and analyses or to publish new information not previously published." Do you need me to clarify anything in that sentence, particularly the second half of it? — HelloAnnyong [ t · c ] 21:12, 28 October 2007 (UTC)

You are being too much of a literalist. Remember the context of the section that edit is in. It is in the "criticisms" section. Therefore the stated fact explains the criticism. While I could problably find a post from some Windows 98 or ME user complaining that he can't play his games after the change in system requirements after scouring the official Steam support forums, I shouldn't have to do so to simply state that obvious fact. -70.232.93.130 21:23, 28 October 2007 (UTC)

Would you consider this to be a good rewrite of that sentence? "By changing the system requirements and making all users run the latest version of the Steam software, Valve is essentially showing preference for users with better computers, and is locking out those who cannot upgrade." — HelloAnnyong [ t ·c ] 17:08, 28 October 2007 (UTC)

That's not entirely accurate, as "better computers" implies above-average. The users who were left out were running systems seven or more years out of date. This kind of thing isn't unusual among software vendors anyway (MS haven't supported Win98 for a long time now, for a pertinent example); the focus ought to be on how the Win98 users had their software forcibly updated and stop working, which is something unique to Steam. --Tom Edwards 18:25, 28 October 2007 (UTC)

So.. Steam autoupdated for Win98 users, and whenever a user would go to play, it would display a message that said something like "You can't use Steam anymore; get a better system"? The solution: find a credible source that demonstrates the behavior you're talking about, describe the behavior in a better fashion, and put that in place of the second paragraph under "Changes to minimum specifications." The paragraph that's there now has to go. — HelloAnnyong [ t · c ] 18:48, 28 October 2007 (UTC)

That's covered in the first paragraph. So long as the second one keeps its current meaning after being cleaned up and being NPOVed, the section will be factually accurate. --Tom Edwards 18:52, 28 October 2007 (UTC)

No, because you have changed the meaning in that rewrite so that it is not entirely accurate and inclusive of the two facts in my version.

"better computers" and "locking out" are too vague. The fact is that the change in system requirements prevents existing Steam users whose PC's do not meet the new requirements from being able to play their previously purchased games. -70.232.93.130 21:03, 28 October 2007 (UTC)

Okay, look. The sentence that is currently there is unacceptable. Rewrite it. — HelloAnnyong [ t · c ] 21:12, 28 October 2007 (UTC)

The original text was: "[The changes] are examples of how the task of ensuring all users have the most recent software can backfire on those with systems that are no longer supported."

An improvement might be: "[The changes] are examples of how Steam's enforcement of most recent software versions can prevent those with systems that are no longer supported from using their once perfectly functional software."

Your change doesn't take into account the fact that changes to the minimum specifications can also be to hardware requirements not just software, e.g. SSE CPU instruction requirement mentioned right above. It's Valve's policy of mandated/forced updates for Steam that makes the changes to Steam's system requirements the issue of criticism since older revisions of Steam can't be used once an update to Steam is released. Also "perfectly functional" is a misleading and subjective term and "software" could be confusing as it doesn't necessarily refer to games which require Steam. "previously purchased games" is an accurate and objective description, and that is how I described it. -70.232.56.5 08:55, 30 October 2007 (UTC)

The text doesn't specify whether the spec change was at a hardware of software level. It only says that it was a software update that caused it. "Systems" is generally taken to refer to the computer in general, if that was what you were getting at. I don't understand why you don't think that the text encompasses the fact that it's a policy of Valve's: Steam is their network and anything it does happens after their decisions. "Previously purchased games" would be fine by me except that not everything on Steam is necessarily bought (e.g. hardware promotions) and not everything is a game (e.g. developer tools). How about simply "Steam-authenticated software"? I'm sure people will be able to put one and one together. --Tom Edwards 12:31, 30 October 2007 (UTC)

Or perhaps to reflect that fact that 98% of the current offerings are games, "Steam-authenticated games and software". --Tom Edwards 12:39, 30 October 2007 (UTC)

As did you and Tom violate it in reverting my comments. It works both ways, see? Tom's having you revert my edits for him without discussion may technically skirt the rule but not its intent.

Our revisions are to the original versions of the text, from before any dispute began. This is allowed and indeed is the proper procedure. Your "comments" belong on the talk page until they can be agreed on. --Tom Edwards 18:25, 28 October 2007 (UTC)

LOL! Your revisions aren't the "original" text. They are simply older than my additions and changes. That doesn't make yours correct or more worthy of use. Despite the fact that you have hovered over this article over the past two years removing edits which conflict with your biased position, you DON'T own this article. I don't need my edits approved by you before I place them , so NO, I will not submit my edits to the discussion page before making them. YOU certainly didn't before you made yours.—Preceding unsigned comment added by 70.232.93.130 (talk • contribs) 20:39, 28 October 2007

If someone wants to dispute existing text, they add the Disputed tag and discuss it on the talk page. If someone wants to dispute a new addition, they remove it (or just add the dispute tag if they prefer) then discuss it on the talk page. This is how Wikipedia works, and if you really want to argue against it you'd better have a chat with this chap. --Tom Edwards 20:04, 29 October 2007 (UTC)

Hey, why am I even doing this? It's not my job to teach people these things. --Tom Edwards 20:18, 29 October 2007 (UTC)

I'd truly like to help in all this, but with all of 70.232.93.130's reverts, I'm completely lost. Also with his comments in here, I don't know what's going on. Scriptdaemon 04:39, 29 October 2007 (UTC)

At this point I've given up trying to sort out Changes to minimum specifications, as it's not actually inaccurate, just confusing, and 70 appears to be in agreement with me on performance effects (or at least, has refrained from commenting on the most recent suggestion while making changes elsewhere). The rest of the activity at the moment is personal accusation and not about the article itself. --Tom Edwards 19:13, 29 October 2007 (UTC)

Cleanup

I'm making this section so we can discuss cleanup on certain parts.

• The gifts and guest passes section is not quite plagiarized from here, but it certainly comes close to it. Might want to clean that up a bit.

I'll add more as I notice things. — HelloAnnyong [ t · c ] 21:50, 27 October 2007 (UTC)

For the sake of accuracy, it should be close to verbatim. Simply properly cite the source I used in whatever format Wikipedia requires.—Preceding unsigned comment added by 70.232.93.130 (talk • contribs) 21:55, 27 October 2007

Please don't put your edits in the middle of others. All I was saying was that we need to be careful. — HelloAnnyong [ t · c ] 22:02, 27 October 2007 (UTC)

This edit introduced a link to a forum. As per WP:RSEX, forums are not reliable sources - so I'm going to remove that edit.

At some point soon, I'm going to go through the article and remove all of the inline links to other pages. One example of this right now is in Steam (content delivery)#Possibility_of_system_failure. These links are ugly and need to be cleaned up. — HelloAnnyong [ t · c ] 18:54, 28 October 2007 (UTC)

How else should we cite public opinion? Relying on news outlets to report on it before we can mention it isn't ethical as that allows those in the media to manipulate, to an extent, what appears on the site. These are not anonymous or newly-registered users - they have verifiable identities, at least within the context of the official forum - nor are they making claims about fact that could be verified in any other way (and indeed if a reputable site did report on the claims, they'd only link to the forum as proof!). --Tom Edwards 19:11, 28 October 2007 (UTC)

There are other ways of garnering public opinion. If sites like Endgadget or Kotaku actually report on something (and not just link to a blog as they so often do) then we can consider using that as a source. As to your mention of ethics, I think that forum posts could and are much more heavily skewed than the media. Forum content is unregulated, so it's unacceptable as a reliable source. For example, if I make a post that says "I have Win98 and Steam runs fine!" and you make a post that says "I have Win98 and Steam won't run!," then we have two conflicting viewpoints, but in theory both would be acceptable. Take a glance over at WP:NOT#OR. — HelloAnnyong [ t · c ] 19:23, 28 October 2007 (UTC)

My point is that these people are reporting their opinions, not verifiable facts. They are criticising Free Weekends as introducing players that they find irksome, and doing so in large enough volumes and with established enough accounts to rule out manipulation (except at an admin level, but that's against the site owners' interests).

Perhaps this is a subculture thing: forum posts in sufficient volume are generally taken as being genuine in the games industry, if not necessarily factually correct. As I pointed out, Endgadget or Kotaku would be using the same forum posts as proof if they reported on it...I don't see how needing them as a proxy for using the posts as a source helps. --Tom Edwards 20:23, 28 October 2007 (UTC)

Fair enough. I'm not really sure what the answer is, so I've asked my question here. We'll see what their answer is. — HelloAnnyong [ t · c ] 20:36, 28 October 2007 (UTC)

Good move. :-) --Tom Edwards 20:37, 28 October 2007 (UTC)

On top of that, the article merely states that the criticisms are being made - not whether they are true or false. --Tom Edwards 20:37, 28 October 2007 (UTC)

Tom, that is the about the only sensible thing you have written here. The criticisms section of the article should be for reported criticisms. Sure, there should be some standards as the criticisms should be at least somewhat popular or widespread, and they should be factually supported or at least logical or reasonable. In addition a counterpoint should be made to the criticism if applicable. The only real problem I have with the criticism section as it is now is that Tom persistently over the past couple of years discourages and prevents new criticisms of Steam from being added (such as the reduced performance criticism) and rewrites the existing ones to lessen their extent and accuracy. -70.232.93.130 21:14, 28 October 2007 (UTC)

There's a response on RS/N; go read it. The person there says basically what I've been saying all along: find reliable outside sources. — HelloAnnyong [ t · c ] 22:15, 28 October 2007 (UTC)

I just have to ask: do both of you want my help editing this page, or do you just want to keep edit warring about nothing? I'd be more than happy to walk away from here if you want me to go. Someone put out a request for a third opinion (which is how I got here) yet it seems that neither of you really want a third opinion. This page is a mess. There are citations missing, the tone is inappropriate, and so on - yet neither of you seem to want to improve this page. All you do is revert each other's edits and then quote each other as to why the previous poster was wrong. You both have some serious reading to do: start with the articles on Manual of Style, verifiability, and citations. Again, I'm more than willing to help you guys build this page up, but if you don't want to do that, or if want me to leave, then I'll go away. Make your decision. — HelloAnnyong [ t · c ] 21:22, 28 October 2007 (UTC)

I have nothing against you personally, but the fact is that Tom called you here simply as a means to get rid of me since he has been unable to run me off as he has done everyone else whose edits conflict with his biased position over the past two years. Tom has no interest in writing the page to conform with anything other than his own bias for Valve and Steam. Tom's previous failed attempt at having me banned and locking this article proves that. As I wrote above, I have no interest in rewriting this entire article myself although I do agree that despite my few edits it is still loaded with Tom's bias and has many omissions on Steam that should be included some of which have been removed by Tom in the past two years. Unlike Tom I have no personal agenda and I am not so wrapped up in this article that I won't walk away. However, I think my current edits should stay as they are unless they become factually inaccurate in the future. Right now they are accurate and balanced in the context where they are placed. -70.232.93.130 21:36, 28 October 2007 (UTC)

"All you do is revert each other's edits and then quote each other as to why the previous poster was wrong."

Excuse me, but I'm keeping highly disputed text off the page here. As a matter of fact I only even began to revert shortly before you (coincidentally) responded to the third opinion request - for the rest of the time I avoided it precisely because I correctly predicted it would only lead to 70 immediately reintroducing his text and removing any dispute tags I might try to add. Anyone reading 70's comment above this one can draw their own conclusions as to what's going on here.

With that out of the way, I'd appreciate it very much if someone who can't be accused of bias could stick around and help. --Tom Edwards 19:30, 29 October 2007 (UTC)

In case you didn't notice, this page was fully protected, so no one's allowed to edit it for the next month. We basically have thirty days to get this page into working order. It seems that there's a few more people around to work on this, so the more the merrier. I'd like to get the entire page revamped into something that's more coherent and close to what this article should be. — HelloAnnyong [ t · c ] 20:21, 29 October 2007 (UTC)

page protected

Since this edit war came to my attention at WP:AN3, the page has been protected to stop the madness. We have a dispute resolution process for things like this. I suggest you make a request for comment on this article, as they seem to get more traffic than 3rd opinions do. There are several steps between an RFC and a request for arbitration, as you recently discovered.

When you have consensus on changes to the article, place {{editprotected}} on this talk page with the content and the place(s) it should go (or the content that should be removed). You can also leave a message for me, but I may not be here as much as the holidays approach, so use the template first. The protection can be removed when the friction stops. Debate good, war bad. Work it out. - KrakatoaKatie 11:04, 29 October 2007 (UTC)

Heh, fair enough. I believe that between Scriptdaemon, Tom Edwards and myself, we can get this page into decent condition. I've only been active on this page for a day or two, but I'm already tired of the edit war. I think we can get this worked out. Should we make a list of the edits on here, or make a page like "Talk:Steam (content delivery)/Protected edits"? — HelloAnnyong [ t · c ] 13:01, 29 October 2007 (UTC)

Holy crap. This article reads like a pages-long rant about why Steam sucks, problems with Steam, past occurrences of Steam glitches, and many, many sections without citations for outrageous claims involving the Steam program. If it were up to me, I'd cut everything off after the section describing Steam's distribution model, and then slowly add back on from there. 24.8.145.162 06:55, 30 October 2007 (UTC)

I'd like to make it known that as the page stands today, nothing is factually wrong outside the sections flagged as being disputed. Uncitable, yes, I admit that, but not untrue. --Tom Edwards 12:35, 30 October 2007 (UTC)

Incidentally, this issue was recently discussed. --Tom Edwards 13:00, 30 October 2007 (UTC)

You're not making a good case for keeping this content in the article, Tom. Uncitable? Strike one. Volume of negative criticism relative to neutral content? Strike two...

Personally, I'd like to take out the "Reduced performance" section altogether, per my comment above under "Fourth opinion". I'd also like to move the Cyber-Cafe Program section to a separate article, as it doesn't relate directly to Steam. Sure, we can mention it here, but the long list of benefits vs. drawbacks doesn't belong in the Steam article. Beyond that, I'd also like to condense the tiny sections into a more cohereent piece. I was about to apply the "toomanysections" tag and work on reducing their number, but that damn revert war got in the way. Nice going, guys. — EagleOne\^Talk 17:47, 30 October 2007 (UTC)

I'm not making a case; there isn't one. I'm just registering the fact that I don't particularly like what's happening. :-) --Tom Edwards 14:19, 1 November 2007 (UTC)

If anything, I've mentioned once before to simply remove all past problems and drawbacks related to Steam. If it's in the past and it no longer applies, I don't see how it matters to the present. Scriptdaemon 01:35, 31 October 2007 (UTC)

I've gone ahead and archived a bunch of conversations on here, as I have a feeling that this page is going to get very big, very quickly. I'll try to get things going in a bit. — HelloAnnyong [ t · c ] 14:20, 30 October 2007 (UTC)

Section cleanup

Okay. Here's a first attempt at getting things going. I guess everyone can post their comments about each section under the header?

Introduction

The introduction seems okay to me. I'd like to add references for each of the clients listed (this for id, etc.) Other than that, it seems okay. — HelloAnnyong [ t · c ] 18:25, 30 October 2007 (UTC)

Client functionality

The Purchases paragraph is fine. I think we should ditch the list of payment methods and gather it into one sentence, something like: "Users can pay via credit card, PayPal, ClickAndBuy, and Visa Electron."

I really don't think anything in the price points section is necessary. It could really be brought together into a few sentences. First attempt: "Games available on Steam are priced on varying levels, where older games tend to be less expensive, and newer releases are the same as the retail price. Prices can also vary depending on country.(reference) Steam has been criticized for the high prices it charges for games added through its publisher contracts.(reference)"

System limitations can go.

Downloads section should be moved up. The following sentences: "A common error in this process is the "content servers too busy" message – which is most likely to mean that the cached "clientregistry.blob" connection information has become corrupt, rather than its literal interpretation.[citation needed] Content is compressed during a download, to a currently uncertain ratio, but unpacked for local storage as it arrives." should be removed.

Merge "Content streaming" with "Downloads", remove the second paragraph.

Compress "Game Cache files" into a sentence or two about how the cache files are used by the mod community.

Merge "Social networking" and "Friends" into a few sentences. H

Integrate "Server browser" into other sections.

Remove "Valve Anti-Cheat."

Thoughts? — HelloAnnyong [ t · c ] 18:25, 30 October 2007 (UTC)

Look good to me, though I'd prefer to keep the system limitations subsection (renamed with a more apropriate title, and trimmed down slightly) because there are some valid limitations. Also, I recommend moving the "Promotions" section out of the "Client functionality" section, because technically those are functions of the Steam service, not the client software. — EagleOne\^Talk 20:32, 30 October 2007 (UTC)

'kay. We'll make that a new section then.— HelloAnnyong [ t · c ] 16:33, 31 October 2007 (UTC)

VAC needs to be linked in the article somewhere. I can't disagree more with removing it...it's a very important component for online games. --Tom Edwards 14:30, 1 November 2007 (UTC)

Sorry, I should've clarified this a bit. It's one line in the article. It doesn't deserve its own section, but I have no problem with integrating it into whatever larger paragraphs come out of this section. — HelloAnnyong [ t · c ] 14:34, 1 November 2007 (UTC)

Promotions

Promotions doesn't need to be broken down. Everything should be merged together and made coherent.

"Gifts and guest passes": First and second paragraphs are okay. Third should be removed.

"Free Weekends" should be merged into another section, and the second paragraph/sentence ("Free Weekends are criticized by existing players[citation needed] for flooding a game’s servers with newbies, and allowing cheaters free reign.") should be removed.

"Hardware promotions": remove self-reference. Remove unencyclopedic tone ("The first (and currently only)..."). Second paragraph also needs to go.

"Statistic gathering": Third paragraph/sentence needs to go. This section either needs to be removed or heavily reorganized so that we can get some citations on it.

"Mods integration": Ditch the second paragraph/sentence. And maybe refactor this elsewhere in the page. — HelloAnnyong [ t · c ] 04:51, 31 October 2007 (UTC)

History

Needs citations all over the place. This also doesn't need any subsections, so to that end, "Early instability" should be worked into the content. "Half-Life 2 Release" is unencyclopedic and either needs to be strongly reworded (that last sentence in particular) or removed. "Steam 3.0" needs to be worked into its own paragraph and heavily edited and cited. Sentences like "...they are believed by industry commentators to be based around the transition of Steam from something that 'just works' to a platform that customers and developers would positively want to use for more tangible reasons than profit margins and independence" need to be cited or removed. — HelloAnnyong [ t · c ] 04:51, 31 October 2007 (UTC)

This is what I was talking about earlier. Do we really need "Early Instability," "Half-Life 2 Release," etc. at all? I personally think it'd be fine just to remove a lot of that. I think the article should focus more on what it is, not what it was. Scriptdaemon 14:20, 1 November 2007 (UTC)

Valve Cyber Café Program

Doesn't need to be its own section. Seems like it should be put under "Promotions." And it should all be reworked into one paragraph, not two lists of bullet points. — HelloAnnyong [ t · c ] 04:51, 31 October 2007 (UTC)

The cafe programme is an almost entirely different system to the normal consumer network. It's got its own client, local content servers, special billing methods, and so on. It should be kept separate in some form or another. --Tom Edwards 14:14, 1 November 2007 (UTC)

Criticism of Steam

Privacy concerns

Remove "The matter of privacy has been raised" and "While it is not widely discussed,". The "To date the only known..." paragraph is already under "Statistic gathering" at the top, so one should be removed. Based on the tone and the subject, I think it belongs under criticism.

Possibility of system failure

Change the tone. The quote seems superfluous and would probably be better as just a reference. The paragraph after the quote should be removed. Last sentence (that references the History) section should be removed.

Resale limitations

A little too close to WP:OR for my liking, though I'd be okay with it if it had more citations.

Forced auto-updating

Citations! Remove the first sentence. Change the tone of the rest to be more encyclopedic. "(Steam is able to stream games..." should be its own sentence. The second paragraph bothers me, but I can't think of what to do with it.

Changes to minimum specifications

Should be its own subsection of Criticism. I've said it before, I'll say it again: the second paragraph is definitely OR and needs to be removed. I understand the sentiment behind it, so maybe it needs to be heavily reworded and factored into the rest. What do you guys think?

Auto-updates and third-party mods

Should probably go elsewhere, maybe under "Forced auto-updating." And it needs to be cited.

Cross-platform Support

Overuse of "some" reeks of WP:WW. I think the second should be squashed into one sentence with a citation (that's not a forum) and moved somewhere.

Reduced performance

Overly simplistic tone. Actually, I think the entire section should be removed.

— HelloAnnyong [ t · c ] 04:51, 31 October 2007 (UTC)

Security

This section should probably be rewritten and merged into the History section. We can just mention that people tried to circumvent the system, but Valve stopped it. — HelloAnnyong [ t · c ] 04:51, 31 October 2007 (UTC)

Localization

Turn into one sentence and merge with "Client functionality" - which should probably be renamed. — HelloAnnyong [ t · c ] 04:51, 31 October 2007 (UTC)

I agree, with the move into "Client functionality" part. I don't think it needs to be renamed. However, I may change my opinion once we start working on the article. Let's hold off on renaming it until we can see the "finished" product. — EagleOne\^Talk 14:19, 31 October 2007 (UTC)

See also

I think we could do without GameTap, GameShadow, EA Link, Stardock Central, Direct2Drive, and RealArcade. — HelloAnnyong [ t · c ] 04:51, 31 October 2007 (UTC)

I disagree. I'm not as familiar with all of those services as I am with Steam, but they're in the same general market. Their pricing models differ (some charge a flat fee for access to the games vs. Steam's pay-per-game model) but there is some overlap. The StarDock and EA Link articles even use the term "content delivery service", just like Steam. In fact, several games are offered on multiple services; take Sam & Max Season One, for example. This was released on GameTap first, then the developer's website, then Steam, and finally retail stores. I think we would be remiss if we didn't include links to those article. PErhaps we could create a List page, like "List of content delivery services" and link to that, a la List of ISO image software, List of LDAP software, or Comparison of virtual machines?

Of course, it's always better to work links like that into the main article, but short of that, we should keep the links in the "See also" section. — EagleOne\^Talk 14:36, 31 October 2007 (UTC)

You're right, they should be worked into the main article, as per WP:ALSO. I just don't want it to be a list where we're listing every service that's similar to Steam. Either a list or a new category would suffice in this case, perhaps. — HelloAnnyong [ t · c ] 05:43, 1 November 2007 (UTC)

Instead of working them into the article, I'd vote for an article on game DD systems in general. It's getting a bit off-topic to start comparing Steam to everything out there on its own page. --Tom Edwards 14:27, 1 November 2007 (UTC)

External links

Remove last two links under "Community." Remove "Other" and its link. — HelloAnnyong [ t · c ] 04:51, 31 October 2007 (UTC)

Whew. Alright, what do you guys think? Again, leave your comments in all the different sections. Once we get some consensus/discussion going, I guess we can start rewriting actual sections. — HelloAnnyong [ t · c ] 04:51, 31 October 2007 (UTC)

Steam Review (and other references)

I just noticed that there are five instances that point to steamreview.org. Tom Edwards has a link to this site on his user page. This presents two problems: one, steamreview.org is not a reliable source since it is a blog, and two, that Tom is potentially skewing this article, and is therefore moving towards WP:COI.

To this end, when we make edits to this page, we CANNOT reference steamreview.org. And it, along with every other link in the Community part of the External links (except for Steam Users Forum) need to be deleted. — HelloAnnyong [ t · c ] 23:13, 2 November 2007 (UTC)

You're going to end with a very short article at this rate. The whole reason I started TSR was because nobody else was covering Steam beyond the press releases! (For the record, I didn't disclose the fact that it's my site before because 70 would only have leapt on that admission and turned up the hate even further.) --Tom Edwards 10:04, 3 November 2007 (UTC)

As per WP:SPS, "blogs are largely not acceptable as sources," with the only exception being when the blog is "produced by an established expert on the topic of the article whose work in the relevant field has previously been published by reliable third-party publications." Sorry Tom, but you don't qualify as that. But hey, if you'd like, I can go put in a ticket at WP:RS/N and ask if the site is reliable or not.

I've been vouched for by the editor Game Developer Magazine a couple of times. Think that'd do it? --Tom Edwards 17:52, 3 November 2007 (UTC)

Fine. I opened a ticket at RS/N. — HelloAnnyong [ t · c ] 18:14, 3 November 2007 (UTC)

How long does this usually take? --Tom Edwards 18:45, 7 November 2007 (UTC)

I don't agree with removing it from External links. Isn't it exactly the kind of thing those sections are for? --Tom Edwards 10:05, 3 November 2007 (UTC)

As far as the external link go, as per WP:EL#Links_normally_to_be_avoided, "Links to blogs and personal web pages, except those written by a recognized authority" are to be avoided. Wikipedia is not a place for you to promote your site. — HelloAnnyong [ t · c ] 15:50, 3 November 2007 (UTC)

Promotion? It's not like I'm running ads or anything. I don't even run a byline. --Tom Edwards 17:52, 3 November 2007 (UTC)

Given that the Steam Review has press level access to Valve, and has broken news such as the Garry's Mod commercialisation, it can be considered a reliable source in regards to Steam. - hahnchen 19:00, 3 November 2007 (UTC)

SteamReview.org has always been fair when reporting news in relation to steam - I would be for keeping all links. --193.1.96.36 13:34, 7 November 2007 (UTC)

Revised content

Okay, I'd like to start putting together revised sections.

Discussion

It's already confusing talking about each section directly beneath it. :-p

On the whole this is a great improvement, and much more readable. But I do think that:

• The paragraphs on validation and VAC are far too short to stand alone. Validation could be merged into another para, and there's more than enough introductory information in the VAC article to expand on its role.
• The Pure server system should be mentioned, as tamper-proofing is a pretty big feature of the GCF format.
• "Weekend or so" can be made more specific: the games that get longer are the ones with longer learning curves, e.g. Red Orchestra.
• The mod integration para needs some touching up now it's lost its heading (I may try this myself in a moment).
• Limitations would read better as a list.

--Tom Edwards 18:03, 3 November 2007 (UTC)

OK, I changed some of the style. I think we may be approaching the Community section from the wrong angle: they're all components of an over-arcing thing, except perhaps the server browser, but they're treated separately. We've got the website, Friends, the game overlay, the notification toasts, the achievements and the stats to squeeze in! --Tom Edwards 18:23, 3 November 2007 (UTC)

There's already an article on VAC, and it's rather comprehensive. What else would you add to the section on here? The pure server isn't even mentioned in the original article. Come up with something to add for it, and we'll discuss. You're right, the mod integration paragraph needs some help. That's why we're here. Propose some edits, and we can talk about them. I do believe that limitations should stand on its own. It can (and probably should) be included in the criticism section; at that point, we could probably turn it into one big paragraph. — HelloAnnyong [ t · c ] 18:20, 3 November 2007 (UTC)

Thanks to an edit collision, see above. :-) --Tom Edwards 18:23, 3 November 2007 (UTC)

Pure servers are the fourth bullet point under Game Cache files. They were in this version too. I made them more prominent. I don't know what I'd add to VAC's mention, but I do know that it looks silly out on its own. --Tom Edwards 18:25, 3 November 2007 (UTC)

I removed some opining from the "distributed file system" section. I also changed the date from 2007-09-12 to September 12, 2007, as the former is hard to read. And I changed fortnight to two weeks and reorganized that section. — HelloAnnyong [ t · c ] 18:41, 3 November 2007 (UTC)

The date thing is a Wiki trick. It'll automatically convert UTC times in wikilinks to the user's preferred format, like this: 2007-09-12. --Tom Edwards 20:25, 3 November 2007 (UTC)

You're right. Sorry about that. — HelloAnnyong [ t · c ] 21:40, 3 November 2007 (UTC)

So.. Anyone have any thoughts? — HelloAnnyong [ t · c ] 20:30, 6 November 2007 (UTC)

Changes to minimum specifications, Forced auto-updates and System failure all relate to the same central criticism: that it's a connected platform which (notwithstanding offline mode) requires connection to and synchronisation with remote servers. It would be excellent if we could merge them in some way to reflect this. --Tom Edwards 15:07, 7 November 2007 (UTC)

Hm, fair enough. I'm not really sure how we'd be able to say that without dipping into OR. What if we made the first line under Criticism "Steam has been criticized for being a connected platform that requires connection to and synchronization with remote servers.(ref)(ref)" and then moved the three categories under it? As a side note: we need references for a lot of statements, or else they're getting deleted. — HelloAnnyong [ t · c ] 18:29, 7 November 2007 (UTC)

Stick some fact tags in and I'll do my best. --Tom Edwards 18:46, 7 November 2007 (UTC)

• Hello Annyong asked me to stop by to take a look, and all I can say is "Wow!" The revision is much more readable (and much shorter), and you're well on your way to a B-class or even a Good Article, which probably wouldn't have happened without the page protection. When consensus is reached, let me know and you can unveil the new, improved version of Steam. Keep up the good work! - KrakatoaKatie 21:23, 10 November 2007 (UTC)

Heh, thanks. Is anyone out there? Tom, other people - do you guys have any further input to this page? — HelloAnnyong [ t · c ] 01:13, 12 November 2007 (UTC)

I'm too busy with assignment work to put any creative thought in right now. I agree with Krakatoa though - you're doing a great job. :-) --Tom Edwards 21:56, 12 November 2007 (UTC)

Very good work, HelloAnnyong. I have a couple of comments. In the intro, we have individual cites for several publishers, and in the payment options section, cites for 3 of the 5 options. Yet, in the Security criticisms section, We have none, not even "cite needed" tags. With sentences like "Although some were true to their claims, the majority were malware," that make specific claims about Valve, Steam, or others, we need to back these up with cites. On the other hand, too many citations in a small space, like with the publishers list, makes the article harder to read.

Also, I vote that we merge the section on the Cyber Cafe Program into the Promotions section. Currently, "Cyber Cafe Program" has all of two sentences. I don't think there's enough content here on that subject to merit a separate section. — EagleOne\^Talk 03:01, 13 November 2007 (UTC)

You're right, the citing is off-balance. I'll shift the intro down to just one citation. As for the rest... well, I'm not really sure what we should do. We can't leave a great deal of uncited information in here. I'm too busy right now to really be able to do research for them, but hopefully someone can get citations in there. If not, I think all the uncited stuff is going to have to be removed. As to the Cyber Cafe thing, that's fine. — HelloAnnyong [ t · c ] 05:06, 13 November 2007 (UTC)

Also, now that I'm looking at it with fresh eyes, I can see at least three instances of the weasel words "Steam has been criticized...". In the first instance, we have a source for the criticism, so we can name the person doing the criticizing, which is preferable to some unnamed entity. The others don't have cites yet, so those will have to wait until we can get more sources. — EagleOne\^Talk 16:33, 13 November 2007 (UTC)

Alright, it's been over a week now, and no one has contributed anything. Are we done debating this page? Are all unreferenced parts going to be removed? What's the deal? — HelloAnnyong [ t · c ] 17:41, 22 November 2007 (UTC)

I went ahead and made a whole bunch of edits. I added a few references in, but I'd like to see some more put in. The block on editing this page is up in a few days, and we need to have something ready to go by then. If no one has any more edits or references to add, then I'm going to have to insist that remove all unreferenced sentences. — HelloAnnyong [ t · c ] 03:54, 23 November 2007 (UTC)

Final Cleanup

Okay, I want to get this page done in the next few days or so, before the block is lifted again. If we can't find references for claims made here, then the claims will be removed. That's how it has to be. So now's your chance to find credible sources and add them in. — HelloAnnyong [ t · c ] 21:47, 29 November 2007 (UTC)

Introduction

Steam is a digital distribution, digital rights management, multiplayer and communications platform developed by Valve Corporation. It is used to digitally distribute and manage games ranging from first-person shooters and RPGs to racing games and cross-genre independent titles. Among its clients are Take-Two Interactive, Eidos Interactive, Introversion Software, Strategy First, PopCap Games, Capcom, id Software, and THQ. As of September 2, 2007 over 200 PC games are available on Steam, and there are approximately 13 million active users.

Comment: Just a first edit. I brought the two paragraphs together, and added sourcing to everything. — HelloAnnyong [ t · c ] 23:41, 2 November 2007 (UTC)

I condensed the citations into one line, as per our discussion. — HelloAnnyong [ t · c ] 03:23, 23 November 2007 (UTC)

Thanks. This looks much better. — EagleOne\^Talk 17:11, 26 November 2007 (UTC)

Client functionality

Steam allows users to purchase access to games through a digital distribution system. Instead of receiving a box, disc, or even CD key, purchased software is immediately attached on the Steam servers to the user's Steam account (which is registered for free), from which it can be accessed and downloaded from anywhere that allows the use of the Steam client. Games can either be bought individually (with some exceptions) or as part of a "package" of multiple games.

The system itself works similarly to a feed reader: The user selects the game they want on their computer and Steam then automates the process of downloading the content and keeping it up to date. The latest version of the game is immediately downloaded, and if there are multiple versions (e.g. a 64-bit edition) the correct one will be chosen automatically based on the computer's hardware and/or software environment. This process happens every time Steam is started online, not just when a game is installed, ensuring that as many users as possible have the latest software. Steam connects over its own internet protocol, independent of the HTTP or FTP protocols used by the web. It downloads only from dedicated "content servers" spread out across the world by Valve and authorised third parties, connecting to several at once to try to ensure a fast and stable connection.

Steam can validate its downloaded content for errors, a process that gives many of the benefits of reinstalling in a fraction of the time.

Steam has a "Distributed File System" that allows a game to launch before it has been completely downloaded. By creating lists of files and requesting them only when about to be needed, a linear game can be begun with only the executable code and a buffer of the first few areas downloaded. In the worst-case scenario, the game will hang while Steam downloads in the background.

Steam-integrated games download to non-compressed archive files with the extension .gcf. This helps to make games more portable, to stop users from overwriting important files, and can be used to prevent files from being tampered with (for instance, the creation of "pure" servers that do not allow custom textures or player models that may give unfair advantage).

Valve Anti-Cheat, Valve's proprietary anti-cheat solution, has been incorporated into Steam.

Steam's interface treats mods in almost exactly the same way as it does purchased games, including, for some, browsable pages on the official site. This is in contrast with most games that offer no built-in launch utility at all. Mods appear in a user's list of installed games with the icons, developer links and other such details that are used by full games. They can also use VAC, Friends, the server browser, and any other Steam feature supported by their parent game. They cannot currently be distributed through Steam however, and as such do not automatically update or use the GCF or NCF file formats. Currently, mods for Valve’s GoldSrc games, Valve’s Source games and Red Orchestra can be integrated.

Payment

Steam's list of payment options, given during each purchase process.

All purchases are made from the desktop through the Steam client, with an encrypted connection, and users are required to enter their billing details from scratch for each purchase as Steam does not store them between transactions. Users are able to pay via credit card, debit card, PayPal, ClickAndBuy, and Visa Electron direct wire transfer.

Games available on Steam are priced on varying levels, where older games tend to be less expensive, and newer releases are the same as the retail price. Prices can also vary depending on country. Gamers have been critical of Steam for the high prices it charges for games added through its publisher contracts.

Edit comments: Per my comments above, I removed one unneeded source (to the "why was my credit card declined?" page), and I moved the other source to the end of the sentence. I think this version reads much better. Also, I changed the edit dates in the sources to the global ISO 8601 format. — EagleOne\^Talk 16:16, 13 November 2007 (UTC)

Changing wording to avoid WW. — HelloAnnyong [ t · c ] 03:51, 23 November 2007 (UTC)

What's "WW"? — EagleOne\^Talk 17:09, 26 November 2007 (UTC)

Weasel words. Sorry, should've clarified that. — HelloAnnyong [ t · c ] 17:37, 26 November 2007 (UTC)

Steam Community

On 2007-09-12, Valve released "The Steam Community" website, a social network that allows Steam users to communicate with each other on a many-to-many scale from both the desktop and through an 'overlay' program within games. Each user's "SteamID" account page contains information such as their Friends (i.e. contacts), how long they have played individual games in the past two weeks, their "Steam Rating" (a 0-10 scale of much overall playtime has been logged in the past two weeks), and of which groups they are a member.

Steam's server browser allows users to search, filter, bookmark and join internet and LAN games for the titles that integrate with it. It works from the desktop and from an integrated game's menu system, and polls Friends to show a list of servers to which a user's contacts are connected.

Friends, Steam's instant messaging tool, supports both one-to-one and many-to-many conversations, held publicly or privately, and VOIP. It provides extended information about what games each user is playing, allowing others to join their contacts in Steam-integrated multiplayer games with a single click.

The Friends system is a popular attack vector for phishers.

Localization

Steam is currently available in the following languages: Danish, Dutch, English, Finnish, French, German, Italian, Japanese, Korean, Norwegian, Polish, Portuguese, Russian, Simplified Chinese, Traditional Chinese, Spanish, Swedish, and Thai.

As a centralized system, Steam also allows Valve to enforce regional lockout on their games. This became an issue when some North American customers bought Valve's The Orange Box from international retailers. The retailers were issued authorization codes specific to their regions. When Valve noticed that these codes were used on computers located outside their regions, the company disabled the authorization codes.

Edit comments: Alright, here's my first attempt at a new first section. I cut out a great deal of proseline, and integrated a bunch of parts into each other. Ten subsections is entirely too much. Anyone have thoughts on what's here? Additions/deletions? — HelloAnnyong [ t · c ] 16:55, 3 November 2007 (UTC)

I added in the "Mod integration" section. — HelloAnnyong [ t · c ] 17:47, 3 November 2007 (UTC)

Just added the localization information in. — HelloAnnyong [ t · c ] 04:32, 4 November 2007 (UTC)

I just added some information on the regional lockout feature, per the discussion below. — EagleOne\^Talk 17:26, 26 November 2007 (UTC)

Promotions

Users who already owned either Half-Life 2 or Half-Life 2: Episode One and who purchased The Orange Box are eligible to give gifts of these games. Valve does not allow these gifts to be bought, sold or traded because doing so violates the Steam Subscriber Agreement, and Valve may disable the Steam accounts of users who are believed by Valve to have done that.

A guest pass is a free, time-limited trial of certain games on Steam that is allocated to a user when he or she purchases an applicable game. The user can then share these guest passes with others who have no purchased the game, allowing the new user to play the game for a limited time. The number of guest passes available to a game purchaser is determined on a game-by-game basis, and guest passes expire one month after being granted if not used. A guest pass allows the user to temporarily share a game license with anyone who has not purchased the game so that person can then play the game for a certain amount of time that varies based on game. Once the guest pass expires, the recipient will be prompted to purchase the game in order to continue playing.

Free Weekends are multi-player promotions in which a game becomes free to play on Steam for a weekend or so. When the promotion ends participant users can no longer play the game, but the game's files can remain installed on their PC's which would save time in downloading future updates if they purchase the game.

Steam has also allowed Valve to run the subscription-based Valve Cyber Café Program, which is the only legal way for a cyber café to offer Steam-based games. There are two pricing models: a flat-rate per-client fee each month or the Valve Time Tracker system that offers a pay-as-you-go model.

Hardware promotions

Steam keeps a record of the hardware in the computer it is running on for various purposes, one of which is enabling hardware manufacturers to run after-sale promotions directly to their customers. The only example of this is the "ATi Steam Offer", which gives free copies of Half-Life 2: Lost Coast and Half-Life 2: Deathmatch to all owners of Radeon video cards, one bundle per card. This promotion began in May 2007.

ATi has also announced on their website that copies of Steam will be distributed alongside all Catalyst driver and software installations throughout early 2008.

Not a very big section, but one to start with. — HelloAnnyong [ t · c ] 17:42, 3 November 2007 (UTC)

I think that the Cyber Cafe program should be part of promotions, since it technically is a promotional device. Tom, if there's really as much to it as you say there is, then you should create an article, and we can link to it from here. — HelloAnnyong [ t · c ] 04:32, 4 November 2007 (UTC)

Meh. --Tom Edwards 15:18, 7 November 2007 (UTC)

Moving the Cyber Cafe program into the main Promotions section, as per discussion. — HelloAnnyong [ t · c ] 03:26, 23 November 2007 (UTC)

History

Steam's development began at an uncertain date prior to 2002. Prior to 'Steam', its codenames were 'Grid' and 'Gazelle'. It was revealed to the public on 22 March 2002 at the Game Developers Conference, and was presented purely as a distribution network. To demonstrate the ease of integrating Steam to a game, Relic Entertainment had created a special version of Impossible Creatures.

The client application, Steam version 1.0, was first made available for download in 2002 during the beta period for Counter-Strike 1.4. At that time, it appeared to be a method of streamlining the patch process common in online computer games. Installation and use of the Steam program was mandatory for CS 1.4 beta testers, but Steam remained an optional component. In 2004, the World Opponent Network (WON) was shut down and replaced by Steam.

Recently, Valve has been negotiating contracts with several publishers and independent developers to release their products on Steam, typically with a pre-order discount of 10% off their MSRP. Rag Doll Kung Fu and Darwinia are two examples, and Canadian publisher Strategy First announced in December 2005 that it would be partnering with Valve for digital distribution of current and future titles.

Half-Life 2 Release

On November 16, 2004, Half-Life 2 was officially released. The game required activation via Steam in order to play the game. While the launch was mainly regarded as successful, later in the day a significant number of buyers (both through Steam and retail) found themselves unable to play the game, due in part to a bottleneck of Valve's Steam system. The European authentication servers went down for about 5 hours before being fixed, preventing those with accounts stored on them from decrypting or playing the game they had bought.

Can't find a reference for the Taylor Sherman quote, so it's being removed. — HelloAnnyong [ t · c ] 03:34, 23 November 2007 (UTC)

Security

Many hacks sprang up following Half-Life 2's launch, each claiming to be able to circumvent Steam and enable the user to get the games for free. Valve responded to these hacks by patching the servers and disabling accounts. It is still possible to download and play some games from Steam, and the games are unrestricted for single-player and LAN play.

Cut out a lot of extra stuff from the history. Scriptdaemon, you mentioned that you didn't like this section before; how is it now? — HelloAnnyong [ t · c ] 20:02, 3 November 2007 (UTC)

Added the security section, more or less. — HelloAnnyong [ t · c ] 04:32, 4 November 2007 (UTC)

Removing the "malware" line, as per discussion. — HelloAnnyong [ t · c ] 03:48, 23 November 2007 (UTC)

Criticism of Steam

System failure

It is necessary to validate every Steam game online before it can be launched, although an offline mode is available. There are no alternate methods of activation such as via telephone or fax, which causes the system to deny access to those without Internet connections. According to the Steam Subscriber Agreement, Steam's availability is not guaranteed and Valve is under no legal obligation to release any such update disabling Steam's authentication in any event.

On the issue of system failure, Gabe Newell, the head of Valve, has said:

“ If you right click on a game in Steam, you'll see that you can back up the files yourself. Unless there was some situation I don't understand, we would presumably disable authentication before any event that would preclude the authentication servers from being available.

We've tested disabling authentication and it works.

”

Temporary system failures, such as in December 2006 when the root authentication servers were unavailable due to storms in Seattle, must currently be weathered by users wishing to play online or make new purchases.

Forced auto-updates

By default, to play a game offline, Steam and the game itself must be fully updated. When Steam starts online, the system checks to see if there are updates available. If there are, the user is forced to wait for update process to finish before being able to play again, though games can be streamed online.) These updates cannot be rolled back by the user, which prevents users with unusual or unrecognized issues reverting their software to its previous, functional state. Steam can be set to stay offline and not attempt a connection,, but this offline mode has its own restrictions and limitations, including preventing games which have not been updated from running in offline mode.

The auto-update model can also affect the large mod community surrounding Steam games. Updates to games have damaged its third-party mods, sometimes to the point at which they cannot run. However, this issue has largely been eliminated with Steam's beta test function, which allows mod makers to self-test their projects with an update before it is released to the wider user base. The process relies on the modders to spot errors, and is as such far from infallible (an issue with one update which crippled the pre-retail Garry's Mod was well-known during the beta period, but not addressed by Garry or Valve until after release), but has in general improved the situation considerably.

Changes to minimum specifications

On June 30, 2007, users who ran Windows 98 or Windows Me were no longer allowed to run Steam or any games that previously supported those operating systems. Additionally, users without SSE processors were warned that Source engine games would no longer function "within the next few months" if they did not upgrade their computer hardware (due to the impending release of its multiprocessor update). However, only a small percentage of Steam users were affected by these changes.

Gaming websites have criticized Valve for not making the system available on Mac OS X. Valve has described the system as as "strictly a Windows application".

Privacy

Steam collects and reports anonymous metrics of its usage, stability, and performance, all without notifying the user at the time of collection or offering an opt-out. Both a notification and an option to opt out are available for personal information. Steam is also used to report similarly anonymous and non-identifying data by several of its games. While some forms of this data are reported back to the public in aggregate form, for instance hardware specifications and gameplay statistics, other non-identifying data has been known to be collected without any indication, as is described in Valve's privacy policy.

The only known example of this undisclosed collection of data has been that of the level of internal fragmentation of Steam's files. The data was used to justify the development of an internal defragmentation utility to reverse the performance-degrading process. "Rather than having to guess or estimate performance bottle-necks", a Steam Update News entry said at the time, "Steam gave us the ability to precisely solve the real-world problem."

Resale limitations

Games bought through Steam cannot legally be resold due to the unsuitability of current Proof of Purchase laws. The only valid proof in an entirely digital transaction is the credit card used during the process, but as ownership of credit cards cannot be transferred, neither can that of Steam-bought games.

Valve does not explicitly prohibit account transfers, but claims that, as a source of fraud within their systems, they are ill-advised.

When a buyer purchases a boxed game he/she must authenticate it with the registration of a CD Key. In the event that the CD Key that they have registered is already in Steam's database, the user is required to submit an image of the physical purchased CD Key for verification purposes, as well as a purchase receipt less than 90 days old. However, if the user has bought used software, either from a vendor, online auction or another user, Valve will not transfer the CD key to the new owner, rendering the software useless.

Payment issues

Despite being available in several countries, prices are always displayed in US$. Users who are not in the United States may also receive an overseas transaction fee from their banks, since all purchases end in an American bank.

Due to Steam products' ephemeral nature and Steam's resultant susceptibility to fraud, billing details entered must match exactly with those held by the bank. Similarly, certain types of failed transactions (such as chargebacks) will cause the user’s Steam account to be disabled until audited by support staff.

Tax is not applied to the quoted price until the very end of the purchasing process.

The last section: criticism. I've cleaned up the sections a great deal - but I've also added a number of fact tags that need to be filled in, or that sentence/section needs to be removed. I moved the limitations section down to criticism and changed the name. Thoughts? — HelloAnnyong [ t · c ] 02:45, 5 November 2007 (UTC)

Reordering as per discussion; also added statement. — HelloAnnyong [ t · c ] 21:51, 8 November 2007 (UTC)

Added two refs. — HelloAnnyong [ t · c ] 03:47, 23 November 2007 (UTC)

See also

• SteamID
• Steam products
• List of Half-Life mods
• Valve Corporation
• Valve Anti-Cheat
• Digital distribution
• GameTap
• GameShadow
• EA Link
• Stardock Central
• Direct2Drive
• RealArcade

External links

Steam website - Official Steam website The Steam Community - web access to Steam's social networking features Valve Software - Valve's official website Valve Developer Community/Steam - Steam category on the official VDC Wiki Steam Users Forums - Official community discussion forums

Added See Also and External links. I removed a great deal of the latter, but we can discuss that. — HelloAnnyong [ t · c ] 04:32, 4 November 2007 (UTC)

Click & Buy

Click & Buy is now a fully supported payment option in steam. The article still states it is in beta - which has been incorrect for the past few months as the beta ended months ago. This needs to be fixed. --193.1.96.36 13:29, 7 November 2007 (UTC)

Price Point

Is it possible to get this edited into the article, maybe rewritten to make it "better".:

Some regions benefit from cheaper games on Steam than in retail because Steam is based on the US dollar. Australian recommended retail price for BioShock is $89.95 AUD ($78 USD), while the game costs only $49.95 USD on Steam. Some regions have certain games or entire distributor collections removed. Australian users are unable to see, purchase or use any THQ game, as prices for THQ games are much more expensive when the user purchases the same game at Australian retail prices than on Steam. These Australian retail prices are often twice as costly than a game converted from US Dollars to Australian when purchased on Steam. Australian recommended retail price for Stalker is $82 AUD ($75 USD), while the game only costs $29.95 USD on Steam ($32 AUD). Macktheknifeau 04:03, 7 November 2007 (UTC)

The page is currently locked, and we're editing it to do cleanup on it. We do have a section about price differences for other countries, but we can certainly add your addition - if you have sources. Without sources, we can't do much. — HelloAnnyong [ t · c ] 04:14, 7 November 2007 (UTC)

Part of this paragraph should be moved to a section about region restriction. --Tom Edwards 15:04, 7 November 2007 (UTC)

There is now a huge array of sources, now that Activision changed overnight, the cost of COD 4 for Australians, from $49USD or so to about $90USD. They also put up the price of Enemy Territory Quake Wars Macktheknifeau (talk) 14:40, 22 November 2007 (UTC)

Regional lockout

The case of recent regional lockout should be added to the criticism section when this lock is over. It received some industry media attention: [1] [2] [3] [4] [5]. --Voidvector (talk) 05:24, 25 November 2007 (UTC)

Wait a minute. The articles claim that some North American buyers bought The Orange Box from international retailers, NOT Steam, to save a few bucks. You can say that Steam was the mechanism used by Valve to lock out those people, but Steam itself isn't to blame. If we include this criticism, it should be carefully worded so that it emphasizes the services role in this incident, and NOT as a criticism of the service itself. I'll bet that those who bought TOB through foreign retailers had bought it through Steam instead, they wouldn't have run into any issues. — EagleOne\^Talk 18:55, 25 November 2007 (UTC)

In simpler terms, Valve used Steam to enforce a regional lockout. If you like, we can add that to one of the Steam features sections. It is a feature like VAC that only penalizes "evildoers" and doesn't affect law-obeying consumers. Also it is a feature that prospective Steam developers may wish to know. --Voidvector (talk) 19:50, 25 November 2007 (UTC)

What makes this unique to Steam? Isn't it true that if you buy an international version of any game, you may or may not be able to play it? What has Valve done that is different from any other software developer/publisher? — HelloAnnyong [ t · c ] 21:49, 25 November 2007 (UTC)

Regional lockout is very rare amongst PC games (although prevalent in console games). Valve was able to enforce a regional lockout using Steam, which would be impossible if the those games were not associated with Steam. --Voidvector (talk) 23:39, 25 November 2007 (UTC)

What's unique here is that Valve was able to centrally enforce the regional lockout through the Steam service, all at once for apparently thousands of users. Even if you buy a Valve game at retail, you still have to authenticate it with the Steam servers. Without Steam, the regional lockout would be much harder to enforce. So, I think this is worth mentioning here, but not in the Criticisms section. This sounds like a feature of the service. If there's one real criticism in all this, it's that the Lockout system may not have been stated anywhere in the license agreement/documentation for Steam. I haven't checked, because I almost never buy games at retail anymore. — EagleOne\^Talk 00:12, 26 November 2007 (UTC)

To reiterate: feature of Steam, criticism of Valve. --Tom Edwards (talk) 11:47, 26 November 2007 (UTC)

Couldn't have said it better myself, Tom :) — EagleOne\^Talk 02:37, 30 November 2007 (UTC)

Reduced performace!

Stop reverting it back, it is common sense that background applications will slow games down. Even Valve tell you that. just read it yourself! Steam uses up RAM, HD space and CPU speed which will have an affect on load times and/or fps. It may have little affect on the games but it will still have an affect.--Worlds best editor (talk) 13:31, 28 November 2007 (UTC)

Wow. The block gets lifted and already people are editing this page? — HelloAnnyong [ t · c ] 13:55, 28 November 2007 (UTC)

Also, neither Valve's page on removing background applications, nor the one on improving game performance specifically mention Steam. Until you can cite a verifiable source that quantifies the effect of Steam on game performance, this section will have to stay out. "Common sense" is not enough for an encyclopedia; you need to back it up with independently verifiable sources. — EagleOne\^Talk 16:58, 28 November 2007 (UTC)

Further Protection

I've protected the page for a further 7 days to ensure consensus has been gained for further edits. Pedro :  Chat  14:15, 28 November 2007 (UTC)

Just for summary purposes: after the protection ran out today, an editor ran in and made these edits. The user changed one section to be a bit cleaner (similar to the new version we have pending here), changed a link to Steam in one section, and removed unreferenced/POV tags in one section. I suspect that this new editor is the same person who was being disruptive before, but was doing so anonymously. — HelloAnnyong [ t · c ] 14:21, 28 November 2007 (UTC)

I have a feeling this guy is going to keep on coming back whenever protection is lifted. --Tom Edwards (talk) 20:55, 29 November 2007 (UTC)

Let's hold off on the speculation for now, and just try to get this page finalized. — HelloAnnyong [ t · c ] 21:45, 29 November 2007 (UTC)

Episode Two Crashes - Nov 21-26

Maybe we can add an example for the forced auto-updating criticisms part. A November 21 update to Half-Life 2: Episode Two caused the game to crash on many parts of the game, and caused a strange "Jumpy" bug in one of the chapters - rendering much of the game unplayable without noclip cheating or savegames, and these methods ruined achievements anyway.

The cause was a memory crash that resulted whenever the inside of certain models' mouths was viewed, because of a "regression" caused by the update. For these six days, this problem persisted and a thread in the Steam forums was populated with over 100 posts, at last ending with a steam update. Sources are Valve themselves, as reported by many support ticket responses, and one thread in the forum in which Valve officials posted the cause and the fix. Also, the update news section of HL2:E2. —Preceding unsigned comment added by 122.52.37.147 (talk) 01:06, 30 November 2007 (UTC)

Sure, we could a brief description of this incident as an example, but details should be left to the HL2:Ep2 article. I never saw this in the game, but I finished it about three weeks ago. — EagleOne\^Talk 01:49, 30 November 2007 (UTC)

List the sources here, and we'll discuss them. Forum posts are not acceptable on Wiki. (see here) It doesn't matter how many posts there are; unless someone reputable comments on it, it's OR. — HelloAnnyong [ t · c ] 04:20, 30 November 2007 (UTC)

The forum post in question was by a Valve official. Additionally, the 100-post thread was closed by a Valve official.

Official statement: http://forums.steampowered.com/forums/showthread.php?t=627619

100+ post thread: http://forums.steampowered.com/forums/showthread.php?t=625858

Within this thread: http://forums.steampowered.com/forums/showthread.php?t=625858&page=25 Post#368 by Valve; last post closes thread and has official statement.

122.52.33.31 18:04, 30 November 2007 (UTC)

Hmm... if anything, only the Valve statement in the last one would be acceptable to use (direct link). This article doesn't need a paragraph on this one little issue; at most it would be one sentence somewhere. — HelloAnnyong [ t · c ] 19:14, 30 November 2007 (UTC)

My thoughts exactly. This was more of a problem with the game, not Steam, so the full details should be left on Episode 2's page. I don't see anything wrong with including one sentence on it somewhere in the article, so long as it's referenced. The auto-update function is really a double-edged sword; it makes it very easy to keep your games up-to-date, and for content providers to push patches out to gamers, but the effects of a bad patch are much more immediate and widespread. All in all, I don't feel comfortable with criticizing Steam for the forced auto-updating. I think we could present it's pros and cons in a more neutral manner. — EagleOne\^Talk 04:12, 1 December 2007 (UTC)

However there is thus the inability to fallback to previous patches, and in fact there is no decision prompt for patch-allow or not. An auto-update system DOES make it very easy, but one with no way to avoid bad patches, unwanted patches, or with no way to avoid the long process of downloading itself, makes the cons outweigh the goods. On forced updating, this is the supreme irritant with which I was faced when reinstalling Counter-Strike Source from a November 2004 HL2 box - spent six hours a day, for a week, downloading patches before I could play. 14:54, 2 December 2007 (UTC) —Preceding unsigned comment added by 122.52.35.145 (talk)

That's more or less the kind of criticism that would be made, but we'd probably go with a single-player game rather than a multiplayer one. You'd have had to have downloaded the patches with or without Steam! The length of download probably wouldn't make it, as the size of games and slow connection you must have are only problems in conjunction with the larger auto-update issue. --Tom Edwards 16:05, 4 December 2007 (UTC)

I'm glad this is protected.

This is perhaps the least neutral article I've ever seen. Almost every sentence attacks Steam (a system that, for the record, I have never used) in one way or another. Until there can be resolution of this immense problem, I see no reason to unprotect in the short-term. -- Mike (Kicking222) 18:10, 30 November 2007 (UTC)

Heh. We're still in need of an admin to help us work out the issues. Both KrakatoaKatie and Pedro have been somewhat involved, but it's been quiet as of late. — HelloAnnyong [ t · c ] 19:10, 30 November 2007 (UTC)

You haven't played Half-Life 2? Unworthy! But still, Steam does have quite a few problems. Some of those sentence attacks may be justified. 02:49, 1 December 2007 (UTC) —Preceding unsigned comment added by 122.52.34.132 (talk • contribs)

What? Where did I say that? Of course I've played HL2. But this isn't the place for game discussion. — HelloAnnyong [ t · c ] 03:17, 1 December 2007 (UTC)

I believe the anon user was referring to User:Kicking222, because you need to use Steam in order to play HL2. — EagleOne\^Talk 04:15, 1 December 2007 (UTC)

Contradition

The section on payment states that "games owned by large publishers tend to be priced at the same level as in stores, if not lower." But under Regional restrictions and pricing (in the critisism section) it says "Regional pricing is widely used by Valve[44] to artificially ensure that prices on Steam stay comparable to or above the retail price of a game in user's area,".

Either the retail value is the minimum price or the maximum, it can't be both. Since I'm not a Steam user (I came hear to find out what it is) I don't know which it is but someone needs to correct that. Danikat (talk) 09:12, 1 April 2009 (UTC)