Talk:IP address/Archive 2

Archive 1

Archive 2

Archive 3

Archive 4

Changes to intro

I changed the intro making it very simple and clear. 0waldo 14:57, 8 April 2006 (UTC) Wow! I suppose that I was totally wrong about that one! 0waldo 15:21, 8 April 2006 (UTC)

---

A comparable analogy to an IP address is a telephone number. On a telephone network, the phone numbers must be unique. Generally speaking, each location has one phone number that allows anyone to dial that phone directly by using their unique phone number. Likewise, a computer is assigned a unique IP address which let's any other computer to connect directly to it by using their unique IP address. In reality, it is possible for a location to have multiple phone numbers (perhaps primarily larger businesses) and a computer can also be assigned multiple IP addresses which is pretty cool.

---

V.S.

---

A computer’s IP address is very much like a telephone number; they are both unique and allow communications with local or remote network devices. Much like a multi-line telephone, a network device may have multiple IP addresses assigned to it.

Consider that a telephone number is "persistent" rather than "transient", and the telephone number, in conventional telephony, to a physical pair of wires. The pair of wires may change due to repair or moves to a new location, but the telephone number stays the same.

A DNS name is persistent while an IP address, if dynamically assigned, can be transient. The closer analogy to telephone number is DNS name, not IP address. The IP address, in turn, may map, on a LAN, to a MAC address. Howard C. Berkowitz 21:46, 25 June 2007 (UTC)

---

I think the latter is better suited but what do I know, anyhow? 0waldo 15:21, 8 April 2006 (UTC)

"local or remote network devices" makes no sense for telephones. Shorter is not clearer. Cburnett 17:04, 8 April 2006 (UTC)

Local and long distance call makes no sense? What about LAN vs. WAN connection? 0waldo 17:40, 8 April 2006 (UTC)

Local and long distance calls are an artificial dichotomy created by phone companies so they can charge for larger network usage. LAN, MAN, CAN, WAN, etc. are also an artificial categorization also for administrative purposes.

Telephone calls between countries, between (North American) area codes, between exchanges, and between lines in the same exchange are quite different in terms of the hierarchical switching model in use. In like manner, the concepts of IP address aggreggation/supernetting versus subnetting are quite distinct. Howard C. Berkowitz 21:51, 26 June 2007 (UTC)

Local or remote phone numbers do not make sense and neither does local or remote network devices in IP addresses. The closest is the localhost but that's entirely different.

Like I said, your shorter version is much less clear because you're using incorrect terminology. Cburnett 18:03, 8 April 2006 (UTC)

Dear Cburnett: I made new and improved changes... Thanks 0waldo 03:00, 11 April 2006 (UTC)

First: a street address is too complicated. Second: a phone number is too complicated. Third: a book index is too complicated. What's next? My opinion of this is that you really want to dumb down this analogy. Each step you change the example to make it simpler but at the same time to the point of using incorrect terminology. This is getting really tiring.

Explaining a technical concept does not require using grunts, chest beatings, and wall paintings. Cburnett 03:14, 13 April 2006 (UTC)

Cburnett: I really liked your last edit; the concept that is, and I made some minor changes that I feel will enable the reader to better understand the basic concept of an IP address via minor syntax changes. 0waldo 04:29, 13 April 2006 (UTC)

Great, but why do you keep deleting the colloquial usage? Cburnett 05:02, 13 April 2006 (UTC)

I removed it because it was not significant; why explain "what an IP address is not" in so many words? Although I do agree with your accurate statement about "IP" being incorrectly used as an abbreviation for "IP address", it is merely not applicable as an introductory item - this is why I clinically deleted it. 0waldo 21:54, 13 April 2006 (UTC)

Cburnett: I removed the "colloquial usage" line. 0waldo 16:42, 15 April 2006 (UTC)

I see that you might compare "extensions" of telephone numbers with the local resolving of shared IP address in a shared hosting environment. but I have never heard nor can i imagine a telephone proxy - in normal terms a telephone number is a GUID that can be shortened by omiting predefined fallback-prefixes (countrycode+areacode) ... now that i think about it - long distance carriers and predial numbers are proxies... got myself - never mind. --Iancarter 09:27, 21 June 2006 (UTC)

Extensions of telephone numbers are a reasonably valid comparison to TCP and UDP port numbers. Howard C. Berkowitz 21:51, 26 June 2007 (UTC)

I think it would be important to list the differences in a Static IP address vs a Dynamic IP address. If I were able to, I would. Can an expert out there add this please?

I'm a little confused here. At least in IPv4, there is no difference between static and dynamic addresses. Whether it is public registered space or private space (RFC1918), it can still be statically configured, or dynamically assigned by DHCP or the IP Control Protocol of PPP.

In all probability, certain IPv6 address ranges are almost certainly dynamic, but, even there, site-local address space could be statefully (DHCPv6) or statelessly (IPv6 autoconfiguration) assaigned. Howard C. Berkowitz 21:46, 25 June 2007 (UTC)

Merge

I see no reason to merge into Internet Protocol. Cburnett 17:13, 8 April 2006 (UTC)

I agree with Cburnett; keep this individual page without merge. 0waldo 18:12, 14 April 2006 (UTC)

Rotating IP addresses

This doesn't appear to present in the article from what I can surmize. Perhaps there can be an inclusion of this somewhere. -Zero^Talk 19:47, 13 June 2006 (UTC)

Could you elaborate on what you mean by a rotating IP address? I've never heard the term used in many years of direct involvement with IP addressing in the IETF and address registries. Howard C. Berkowitz 21:52, 26 June 2007 (UTC)

I get the feeling he means DHCP. In that case, DHCP. - seandon mooy

Yes. Most of the dial up service providers provide DHCP. Even individual cable line providers can also assign DHCP to their clients. Some ISPs assign static IP to clients on request or on the pay of additional cost. Kumarsarma (talk) 23:26, 28 December 2008 (UTC)

If you are client of local internet cable service provider, you need DSL or ADSL routers that can be provided by your ISP and static IP can be assigned to it. Kumarsarma (talk) 00:03, 29 December 2008 (UTC)

What is so special about whatismy?

I feel that some editor abuse again wikipedia!? What is so special about http://whatismyipaddress.com/ to be allowed above all other websites that are much much better. Furthermore site have not only static advert but also a couple annoying advvertisement (as pop-up that jump every time when you visit site, then annoying blending and scrolling advert in middle of screan etc...) Arrrggg......... Only reason could be that some of wikipedia editor use his power and simple allow his own site. Here was a discussion about years a go and that time everybody agree to remove any single ip address displaying site except articles about tcp/ip protocols on well known sites. Hmmm........ Looks to me as abusing of power again. —The preceding unsigned comment was added by Graciella (talk • contribs) 19:10, 3 August 2006 (UTC).

The website specified by you may not show accurate results in look up and tracing if your connection is DHCP client or connected to Proxy server. I believe that members who posted information in this page didn't explain clearly about static IPs, DHCP and proxy connections. —Preceding unsigned comment added by Kumarsarma (talk • contribs) 01:34, 29 December 2008 (UTC)

You do realise that we're all editors here on Wikipedia? In this case it seems to be this edit by Gracefool (talk · contribs) which added the whatismyaddress link over two weeks ago. Odd, but he doesn't look like a linkspammer. Anyway, if you have so much trouble with popups, I suggest you switch from using Internet Explorer to view web sites. --Imroy 19:23, 3 August 2006 (UTC)

I guess that you didnt get my point because of your last sentence with suggestion for popups.

I do not have personally problems with popups but have problem that someone post link to only one site (that has been chosed above all others and have obviosly special status) and generate a huge traffic to that site. In fact the site in question is advert for free on wikipedia and in the same time administrator make his own site ugly with annoying popups and scrolling advert in middle of screen and generate money etc). As first why to place link only to one site with a lot popups and advertisements? The site does not have anything special. Administrator, friend, neighbour or whatever could be reason but visitor friendly, educative and usefull site not for sure... Wikipedia is not place to place advertisement and it was advertisement. Graciella

It's hard to understand what you're getting at. You seem to be annoyed that someone added a link to an ad-laden website, complete with popups and whatnot. That I do understand. I often revert "link spam" being added to articles. But I think you're misplacing the blame or not understanding how articles are edited. None of these changes are approved in any way. The sole blame lies with Gracefool (talk · contribs) and noone else. He/she made the edit. The fact that the link remained for about two weeks is a mere oversight. It happens. Thankyou for fixing it, now move on. I don't understand why you are making such a huge fuss over it. --Imroy 12:49, 4 August 2006 (UTC)

International IP ranges

It would be interesting to have some information about this.

I know that a portion of an IP number carries information about a user/website's geographical location/country of origin. —Preceding unsigned comment added by 200.142.100.230 (talk • contribs) 19:24, October 4, 2006

The address itself does not carry any such information. IP addresses are assigned by a central authority, IANA. A certain range may have been assigned to a company in a specific region, and it is possible to look an address up in IANA's database to see approximately where it was assigned to, but the address itself is just a number. 02:52, 10 February 2007 (UTC)

Actually, IANA delegates to regional internet registries (RIR), which are roughly continental. Nevertheless, it's not practical to take more than a guess at an IP address's physical location. Addresses starting with 47.x.x.x are assigned to Nortel, headquartered in Canada. A Nortel corporate location in Japan could very well be in that range.

Simplifying a bit, you can infer the organization that had the address space assigned from IANA or a RIR, but when that organization is an ISP, it will subdelegate parts of the address space. The WHOIS and SWIP databases may or may not reflect that delegation in a geographically meaningful way. Howard C. Berkowitz 21:56, 26 June 2007 (UTC)

Vandalism

"An IP address (Internet Protocol address) is a unique penis that devices use in order to identify" someone fix this plz. —Preceding unsigned comment added by 71.57.127.191 (talk • contribs)

"xcxcxc" right at the start - I assume that's not meant to be there. It doesn't appear when editing a page so I can't remove it myself. --Edvvc 12:54, 1 March 2007 (UTC)

"in simpler terms, a computer address. An IP address can easily be used to enter your computer through a back route, enter your home through the same back route, enter your brain, obviously, kill people, download tickets, illegally bypass blocks, and cause your life to crumble down into a tiny pile of horse crumbs" -- vandalism —Preceding unsigned comment added by Landcamera900 (talk • contribs) 14:35, 12 December 2007 (UTC)

Willie Pep

I believe that the 'Ray Famechon' you mention as having fought Willie Pep in 1950 was a Frenchman, not an Australian. It is his nephew 'Johnny Famechon' who is Australian, but I believe Johnny fought in the 1960s.

~ Jaimee K Pimentel, Sydney, Australia

Recent External Link Additions

(Moved the following entry from a comment on the main page —Krellis 16:19, 24 December 2006 (UTC))

Hey guys, attempting to post this link again as a registered user (deleted by Krellis last time). This site is 100% free and ad-free site that definitely doesn't just show "dime a dozen" information about "your IP address." From a single page - you can find out geographic location, ASN/BGP information, route, whois, network peers (using BGP feeds), and a lot more. I believe this is a VERY helpful, "hands on" service for someone wanting to know more about an IP address. If you still feel this isn't relevant and is removed, I will not post again. Thanks, -Josh —Preceding unsigned comment added by JoshSkidmore (talk • contribs) 04:39, 24 December 2006

Actually, I didn't remove the lookupcrap.com link, I removed one to ipphoto.org, but I don't really think either belongs there. Sure, it does show a lot of information about an IP address, but the comment I referenced still applies - there are a million sites out there that provide that kind of information. Additionally, while the name "Lookup Crap" may be kind of amusing, it isn't exactly very polite or professional. I personally do not think it is appropriate, but I'd like to try to open up discussions here and see if there's any consensus. Other thoughts? —Krellis 16:19, 24 December 2006 (UTC)

Krellis, I agree that there are many tools out there similar to the one I've listed. Even though you've decided that the tool is not appropriate for the "IP Address" wikipedia article, I do think that a "hands on" tool similar to this needs to be listed in the "External Links" section of the article. I could almost guarantee (if polled) that 8 out of 10 people that land on this article would love to use a tool to learn more information about an actual IP address. But again, I do understand that this article must get a LOT of spam-like sites that tell about an IP address (surrounded by 400 ads and revenue methods). As for the name - the word "crap" is a word that's in almost everyone's vocabulary, and it though some people would rather not use it, it's not considered a "curse word" and more than likely does not bring offense to anyone. The name will stick, and if you ever need to get back to the site (for whatever reason) in two months, you'll remember the name :). (I have also reverted Graciella's revert since they believed it to be spam since there's no comment.) —JoshSkidmore 18:52, 26 December 2006 (UTC)

If you look at the archive of this talk page, you'll see a number of discussions where, at least from my reading, consensus has already been formed that this type of link should be avoided altogether, to avoid any need for judgment calls of which site(s) are okay vs. which are not. Specifically, note the Vote for "show your IP pages" section. I also note from WHOIS information that lookupcrap.com is owned by one Joshua Skidmore, presumably you - that is something that you really should have mentioned from the outset. Does anyone else want to weigh in on whether or not this site should be included? If I haven't heard anyone who wants to keep it (other than its creator) in a few days, I'm going to go ahead and remove it, and ask that it not be re-added. I'm just not convinced that there's sufficient additional content to make it a valuable addition to the article. —Krellis 19:54, 26 December 2006 (UTC)

Krellis, thanks for the heads up on this. I really was very unaware that there is this much politics behind Wikipedia. I'm going to go ahead and remove my link. An idea I was thinking about (and might have been mentioned before): Why not have a new listing article of "IP + Hostname Tools" in which all of the services can be listed? In a straightforward fashion, the community can bluntly list the features of each site (eg: "limited to 10 lookups per day," "does XXX feature not found on other sites," etc). I can understand where a lot of people get annoyed by these links, but at the same time, some of the tools can be very useful to people, and it's a shame this article's editing community has decided to remove all links altogether. I can start this page? —JoshSkidmore 21:00, 26 December 2006 (UTC)

(De-indenting for easier reading...) You certainly can start such a page, but I'm not sure it's really appropriate for Wikipedia - see WP:NOT for a list of what Wikipedia is not intended to be, including "a mirror or a repository of links, images, or media files" (section 1.5 of that page). Lists of this type already exist, particularly in the form of Google Directory. In particular, the DNS -> Web Tools category seems to be similar to what you're suggesting, as is IP -> Addressing -> Lookup. Perhaps you would be better served by using the links at the bottom of those pages to help build them up (particularly the latter, which is somewhat empty), and then linking to THEM from here on Wikipedia, as links to relevant web directory categories are generally considered acceptable under WP:EL. I don't really think it's about politics, as you put it, as much as it is about making sure Wikipedia remains a valuable resource, and that means taking steps to ensure it doesn't become a cluttered mess of links. I understand that it may not seem that way to you, but I hope my suggestions here have made it a little bit more clear and easier for you to understand. —Krellis 21:16, 26 December 2006 (UTC)

I appreciate the comments and I agree with you completely. I didn't mean for the comment about Wikipedia being all about politics to be taken with a negative connotation, but the history pages do have a lot of people similar to me posting links and then arguing with editors on why or why they shouldn't be listed on Wikipedia. I am also sorry that it may appear that I'm just trying to use this as a means to promote my site, but a lot of the reason I created it was to "extend" knowledge about the core aspects of the internet (often listed here). I didn't see my site as spam, but I can definitely see how adding the site to "External Links" on articles can be seen as that. Bottom line here is that I am new to editing Wikipedia, and I should have be a little more cautious. Hopefully I help by contributing content where it's needed. Thanks again! —JoshSkidmore 21:32, 26 December 2006 (UTC)

check ip

check ip —The preceding unsigned comment was added by 200.171.68.160 (talk) 11:40, 23 January 2007 (UTC).

I edited the page to remove some useless information. I simply reverted (I guess, Im new at this) the page back to a previous state. I did that without logging in so I just wanted to sign this and let you know what I was doing. Thanks. Illumini85 01:51, 6 March 2007 (UTC)

IPv5 copyright issue?

I noticed that the bit on IPv5 being experimental and referring to UNIX numbering conventions is lifted almost verbatim from whatismyip.com (near the bottom of the page). There is also an external link at the end of the article (also possibly shouldn't be there, per other discussions) to a whatismyip clone (ip.cyarena.com), containing the same text.

On neither website is there an obvious claim of copyright that I can see, but does this automatically allow us to copy from the website? JoeKearney 09:25, 23 March 2007 (UTC)

Is this even worth mentioning? IPv5, IPv7, and IPv8 were all alternatives being considered for the Next Generation IP to replace IPv4. The IPv6 proposal won. Hcberkowitz 21:49, 25 June 2007 (UTC)

Protect?

Should this article be protected? Since I started watching it I've noticed a large amount of vandalism, and there don't seem to have been any constructive edits for a long time, so I don't think it will hurt. Since it's linked from vandalism warnings etc it's an easy target, and without protection a sitting duck. Richard001 00:13, 14 May 2007 (UTC)

It looks like the level of vandalism is a lot more acceptable now, however I don't think going back to unprotected is an option. Already we see vandalism appearing on the talk page from frustrated users who can't edit the main page. The ban will be lifted on May 29, but if vandalism returns after that, which it will, it looks like it will be indefinitely protected. Richard001 00:13, 18 May 2007 (UTC)

Things aren't looking any better after unprotection - we're just getting the same garbage day after day. Does anyone have any objection to permanent protection? Richard001 09:23, 7 June 2007 (UTC)

Address divisions

I was wondering if anyone knew the signifigance of each of the four divisions in an IP address. I only know that the first six numbers seem to stay the same for a general location, like an office. I think the article should have this information. Redneck Physicist 21:39, 24 May 2007 (UTC)

Each IP address is four octets long, so it is usually expressed as four numbers with dots, but it is really just a number which is 32 bits long. There isn't an inherent meaning to any chunk of the IP address, but it can generally be considered as being divided in two parts. The first part is the address of the network, and the second part is the address of the host on the network. But, the length of the part which specifies the network isn't fixed. Since the specific octets have no specific meaning, I'm not sure what information you want included, exactly. Forkazoo 17:07, 25 May 2007 (UTC)

It's the netmask that defines which portion is the network address and which portion is the host address...there's now way to tell by viewing only the IP address (although an educated guess could be made based upon the class that the address belongs to.) In order to explain this here, binary and ANDing would have to be explained as well.71.196.245.252 21:32, 12 June 2007 (UTC)

Even address classes went away, a long time ago, with Classless Inter-Domain Routing. Remember that IPv4 addresses are 32 bits long, and the mask can conceptually be of any length, not on octet boundaries. The newer notation states how many bits are used in the prefix, such as 192.0.2.0/24 being equivalent to 192.0.2.0 with mask 255.255.255.0. In like manner, 172.16.0.0/12 is equivalent to 172.16.0.0 with a mask of 255.240.0.0. Hcberkowitz 21:53, 25 June 2007 (UTC)

Static vs Dynamic

I think that this entire section should be removed completely. An IP is an IP is an IP...doesn't matter where it came from. This is like saying there are two types of cars: the kind you buy or the kind you lease.

Additionally, this stub is misleading and inaccurate. For example, a static address can also be obtained via a DHCP server, not just dynamic addresses (ie reservations). This sentence: "With a static IP address, a computer's identity can be easily identified by others, and users can easily connect with it" implies that a dynamic address is difficult to identify or connect with...not true This is what DNS and dynamic DNS were created for...71.196.245.252 21:27, 12 June 2007 (UTC)

Another thought: a better way to "classify" IP addresses would be public vs private. I don't see this mentioned here and is a much more appropriate way of grouping types or addresses than static vs dynamic. thoughts?71.196.245.252 21:37, 12 June 2007 (UTC)

I am proposing to keep the section and I also propose (and welcome revisions to) the following:

---

begin proposed changes ------------------------------

There are two classifications for the allocation and assignment of IP addresses: static or dynamic. A 'Static' IP address is generally assigned by a person responsible for the network but a simple user can also assign a ‘Static’ IP address. 'Static' IP addresses generally do not change once assigned and are generally assigned to permanent devices connected to a Local or Wide area network such as web and mail servers where a large group of users access that network device.

Contrast the manual or ‘Static’ assignment of IP addresses with the “Dynamic” method where the addresses are assigned automatically by a special protocol running on a host computer or router; that protocol is DHCP, which is an abbreviation for Dynamic Host Configuration Protocol. An example of automatic assignment via DHCP could be: a person takes a laptop to a wireless zone at a coffee shop, the settings for IP protocol on that laptop are set to ‘obtain an IP address automatically’ – the laptop broadcasts a DHCP request over the wireless network and if a DCHP server is found then an IP address is assigned to the laptop for a predetermined period of time – this is ‘lease’ period and this duration may range from minutes to infinity as pre-defined by a network administrator for that network. Dynamically assigned IP addresses are the preferred method of assignment as the DCHP program does the majority of the accounting and administration thus ensuring that no duplicate IP addresses are assigned on the network

---

end proposed changes -------------------------------

sorry, but you are confusing the IP address itself with the means by which it is assigned. DHCP is decidedly not preferred for all addresses. The trivial example is that if everything is DHCP assigned, how does the DHCP server get its address? Further, DHCP is not the only way addresses are dynamically assigned in IPv4. The IP Control Protocol of PPP also can do dynamic assignment. In some implementations, IPCP may get information from a DHCP server, but it often does not. Howard C. Berkowitz 22:00, 26 June 2007 (UTC)

what are you talking about? 0waldo 13:47, 27 June 2007 (UTC)

Internet Service Providers charge extra for static IP addresses. Why would I want to pay extra for a static IP address? What can I do with a static IP address that I cannot do with a dynamically-assigned IP address? Csnewman 19:45, 23 July 2007 (UTC)

There are several situations where a static IP address is desirable, although there are many others where people want one not because they have to have it, but because they are accustomerd to having static addresses when they do system administration.

The most basic is for servers or routers that will need to be accessed through the Internet, perhaps indirectly through a DNS lookup. DNS clients (using the term a little loosely) cache the value they obtain, so if lookup of the server foo.example.com returned the address 2.3.4.5, the application client that requested the address lookup would send to 2.3.4.5. If the address were dynamic and reassigned, the server might now have the address 192.0.2.123, yet the application client would still send to 2.3.4.5.

A more technical reason involves routers that learn the addresses of other routers through routing protocols, which do not use DNS forward lookup at all. If my Internet gateway router believed that a given destination address 172.16.31.9 were reached by forwarding to the next hop 10.27.33.96, but the destination address were dynamic and changed to 192.168.33.86, my router would send to what it believed to be the correct next hop, but that would be taking the wrong "road" to the destination. A less technical reason, when dealing with routing, is "much magic. Gods want it to work that way." Howard C. Berkowitz 21:18, 23 July 2007 (UTC)

Class C Address Range?

Please check the Class C private IP range example in the article It is given as shown below: C 256 = 28 192.168.0.0 192.168.255.255 But i feel the end address should be 192.168.0.255. Am i wrong? —Preceding unsigned comment added by 203.158.89.10 (talk) 05:59, 10 December 2007 (UTC)

Yes, the table in IP Address#Private Networks is correct. Check the RFC listed in that section. The confusion comes from the fact that the 3 listed networks aren't example networks, they're private network blocks reserved by IANA. I've made some changes to the wikitable; specifically, I think having the network class in the first column (and in bold with a darker background) made it look as though the table illustrated examples of the 3 classes. The names I listed for each network block look pretty awkward, but they're the ones used in RFC 1918.

I also removed the "Number of Possible Networks" column, for two reasons: I think it only makes the table look more complicated, and I think it is confusing since (as RFC 1918 mentions) it's only relevant in a pre-CIDR environment, which most readers will not necessarily be familiar with.

I'm open to different ideas, though. Indeterminate (talk) 04:49, 18 December 2007 (UTC)

ipconfig used to assign a fix IP address?!

in the article is written "Static IP addresses are manually assigned to a computer by an administrator, either through the operating system configuration or through a command (e.g. ipconfig or ifconfig)" I think is wrong. As far as I know ipconfig is used to display the ip configuration (as is written in the related article). --Dia^ (talk) 18:21, 9 December 2007 (UTC)

Since no une is giving any answer, I'm going to correct the info.

By the way all what can be dame with an ipconfig is:

USAGE:

   ipconfig [/? | /all | /renew [adapter] | /release [adapter] |
             /flushdns | /displaydns | /registerdns |
             /showclassid adapter |
             /setclassid adapter [classid] ]

where

   adapter         Connection name
                  (wildcard characters * and ? allowed, see examples)

   Options:
      /?           Display this help message
      /all         Display full configuration information.
      /release     Release the IP address for the specified adapter.
      /renew       Renew the IP address for the specified adapter.
      /flushdns    Purges the DNS Resolver cache.
      /registerdns Refreshes all DHCP leases and re-registers DNS names
      /displaydns  Display the contents of the DNS Resolver Cache.
      /showclassid Displays all the dhcp class IDs allowed for adapter.
      /setclassid  Modifies the dhcp class id.

The default is to display only the IP address, subnet mask and default gateway for each adapter bound to TCP/IP.

For Release and Renew, if no adapter name is specified, then the IP address leases for all adapters bound to TCP/IP will be released or renewed.

For Setclassid, if no ClassId is specified, then the ClassId is removed.

Examples:

   > ipconfig                   ... Show information.
   > ipconfig /all              ... Show detailed information
   > ipconfig /renew            ... renew all adapters
   > ipconfig /renew EL*        ... renew any connection that has its
                                    name starting with EL
   > ipconfig /release *Con*    ... release all matching connections,
                                    eg. "Local Area Connection 1" or
                                        "Local Area Connection 2"

--Dia^ (talk) 15:00, 13 December 2007 (UTC)

What problem are you trying to solve? Howard C. Berkowitz (talk) 22:21, 17 December 2007 (UTC)

Oops! You're quite right, Dia^. I did that. I'm so familiar with ifconfig, guess I just forgot that ipconfig doesn't actually let Windows users change their IP address. I apologize, thanks for catching my mistake! :) Indeterminate (talk) 05:12, 18 December 2007 (UTC)

Moved "IPv4 Specifics"

I moved the section "IPv4 Specifics" into the "IP version 4" section. I tried to keep most of it, but this is ridiculous. This article is for IP Addresses. If you want to add information on networking, subnetting, IP blocking, NAT, or any other IP-related subjects, those topics are usually important enough that they have their own articles. Please try not to clutter up this article with information that's more relevant elsewhere. Indeterminate (talk) 06:17, 18 December 2007 (UTC)

Trailing /xx on IP addresses to specify ranges?

I came to this page to find out what this means, but although the page uses it in a number of places, it never explains what it means! Thanks 131.111.195.8 (talk) 18:15, 31 May 2008 (UTC)

It is part of the CIDR notation, but I notice it is not well explained on that page either. At least not without reading a lot of the article. Looks like both pages could use some rewriting to make it easier to find answers such obvious questions. Wrs1864 (talk) 18:28, 31 May 2008 (UTC)

Ah, thank you. I think that for now it would suffice to simply say somewhere that it represents the number of bits from the left of the IP address to match to, and then a link to the CIDR page. Not sure where to put it... 131.111.195.8 (talk) 12:52, 1 June 2008 (UTC)

I added a brief explanation under the Subnetting section. If anyone wants to expand it, go ahead. Indeterminate (talk) 20:04, 1 June 2008 (UTC)

Unsourced Statments

I have tagged and highlighted two unsourced statements in the opening paragraph of the article. I suggest that if ref's cannot be found to back up these statements, then they be removed from the article. Brothejr (talk) 12:00, 24 June 2008 (UTC)

Thanks. They were incorrect and have been removed. Indeterminate (talk) 20:38, 24 June 2008 (UTC)

Recent edits: early July

So I notice most of the page has been rewritten over the past few days (mostly by Kbrose). While I agree with most of the revisions, I do have a few small complaints (and I've made some changes):

• The introduction was quite long. Seven paragraphs makes for some dry reading before getting to the table of contents. This isn't a book on IP addresses. I pared it down to 5 paragraphs, but it still seems long. Can we try to keep it around 3-4? I'm sure we can come up with a simple, easy-to-read introduction to IP addresses that briefly covers the basic sections of the article.
• I tried to integrate paragraph 2 of the intro into the rest, omitting errors. For instance, IP addresses should be unique (on a network), but they are not necessarily. See below. Also, I think saying that they identify a computer is redundant. I liked the simple, clear wording of this paragraph, however.
• I also integrated paragraph 3. The Internet Protocol is the primary network layer that deals with IP addresses. - This sentence was confusing. IP is not a network layer, it's a protocol in the network layer, and of course it deals with IP addresses, that's why they're called IP addresses. Elephant trainers also primarily train elephants. :)
• After looking through RFC 760 and RFC 791, I was unable to find any statement that IP addresses should be unique to a network node. So I added a Fact template to the statement, until a suitable reference is found.
• (from the IPv4 Private Addresses section) "Three ranges of IPv4 addresses for private networks, two per IP address class (A and C), and a CIDR prefix within class B, were reserved in RFC 1918; these addresses are not routed on the Internet, and thus need not be coordinated with an IP address registry." I can almost understand why the bold section was changed from the previous entry ("one per class"), but this was incorrect and unclear, so I changed it. The private nework range in class A was an entire class A network block, but neither the 20-bit or 16-bit blocks were entire B or C class blocks. I prefer the simpler phrasing of "one range for each class", since that appears to be how they decided on the number of private network address ranges.
• I think we still need more citations.

In conclusion, I think the introduction (and much of the rest of the article) is still too technical for a general audience. If you guys disagree, let's discuss it here! :) Indeterminate (talk) 12:24, 8 July 2008 (UTC)

response

I think the intro should be sufficient to be used as a stand-alone definition of the topic, so that if someone does not want to read the whole thing, they have a good representation of the topic, in that sense I think the intro isn't too long as it stands. It shouldn't be so short and abstract that it provides no meaning full insight or satisfaction even for a knowledgeable reader, and technical content shouldn't be too diluted, it is a very technical subject after all. I think the changes made to my version do make it more concise, less verbose and are an improvement. I think there is more that can be done, but I don't see that it could get shorter or less technical. Google provides enough simple-minded answers already. No need to come here. My revision was intended as a first iteration.

I added the routing paragraph for one reason: People constantly ask (me and elsewhere) what the CIDR (/16) designation means ("haven't seen that kind of IP address before"), and it may well be the single reason a lot of people look. That's the reason I added the paragraph on routing, to be able to clear up that issue as early as possible in the article. You removed the most important part of that paragraph, I think. Without the example CIDR notation, the routing paragraph might as well be shortened radically, since routing has its own page. I would like to reintroduce the CIDR notation in that paragraph, but in a very succinct manner.

Regarding the "primary network layer" phrase, it was intended to be understood to indicate that IP is the *PRIMARY* layer that deals with IP addresses, as many layers in the protocol suite do deal with them of course. But the layer that introduces them to the stack is IP. Some people think that the network layer consists of multiple layers itself, I intended to used the term more independently. It wasn't very clear afterall, I agree now. I don't want to get into the layers debate, I am just glad the TCP/IP layer template isn't attached to this article as well.

It is true that there just aren't good definitive references for the concept of uniqueness of addresses. It seems people keep adding stuff on uniqueness of addresses, there was another one overnight. It was in the first sentence of the old intro. The topic was good for the new paragraph as it leads into private networks. The concept of uniqueness can't really be attached to the general term IP address, since every IP device has 127.1 localhost address. IPv6 addresses certainly can't be defined as unique even on the same network, but in the case of IPv4 it does need to be unique on the same wire. That's why I tried to phrased it the way I did, since ARPANET back then wasn't very subnetted, and the concept of uniqueness was somehow implied if only simply in the process of downloading the host table and installing it on your host. I struggled with the topic at the beginning of that paragraph very much, and I still wanted to add better handling of it. I like your addition of mentioning NAT at that point, good to link to that article from here.

The rest of the article is still in need of overhaul, IHMO. Kbrose (talk) 15:50, 8 July 2008 (UTC)

Re: Re:

I guess you're right about the length of the intro. At least, I can't decide what to get rid of. I just look at articles like Binary numeral system, and the tiny introduction lets me just glance at the table of contents and skip to the section I'm interested in. *sigh* Anyway, I do really like that quote from RFC 791. It might be the best part of this article.

I agree about CIDR notation, too. I know it's confusing to a lot of people. I added it back in; if you want to reword it, feel free. I've also been thinking about adding a short section above IP versions to mention the different notations used for IP addresses: dotted decimal, hexadecimal, binary, IPv6, as well as CIDR and ip/netmask pairs. Maybe I'll try making an image which has one example of each. Can you think of any other, uh, notable notations we should mention?

Yeah, the layers stuff is a mess. I watch Internet protocol suite too, and despite the fact that everyone agrees that the TCP/IP layer model template shouldn't have 5 layers, nobody's changed it. People are very opinionated about some strange things.

Mentioning uniqueness is tricky. Like you said, they only implied that addresses should be unique on a network in the old RFCs. And especially if you take IP address spoofing into account, it's obviously possible to have addresses that aren't unique on the same wire (although presumably not intended by the protocol spec). Loopback is an interesting case, too.

There are definitely some changes I'd like to see happen in this article. Most of this article is unfortunately just a collection of brief descriptions of other main articles (IPv4#Addressing, IPv6#Addressing, Subnetworks, Private networks, NAT, etc.). But I don't think there's really anything we can do about those, aside from cleaning up the text. The only section that seems to be unique to this article is just the static/dynamic addresses section. Like I said, I'd like to write a section on notation, and we should probably mention Loopback/localhost addresses, network addresses (.0), and broadcast addresses somewhere, since they are special. Anything else you can think of that's obviously missing?

Anyway, thanks for responding. I'm glad you're helping out here. Indeterminate (talk) 01:11, 9 July 2008 (UTC)

RE RE RE: Thanks for putting the CIDR back in. I'll think more about further changes. RE address spoofing: of course it's always possible to put anything that fits in an IP packet, but you can't count that against uniqueness of network interfaces addresses on the wire. cheers Kbrose (talk) 13:36, 9 July 2008 (UTC)

Cool

Good Work —Preceding unsigned comment added by 72.138.43.61 (talk) 22:10, 20 April 2009 (UTC)