Wikipedia:Featured article candidates/Microsoft Jet Database Engine/Archive1
Self nomination (one of my focus articles). I think this is pretty comprehensive now! I have referenced it thoroughly, and attempted to explain the various database concepts as well as I can to the layman (one of the peer review comments). I have also got a table that details versions for the history - I realise that the history text is quite dry, but nonetheless necessary. Hope to have this up to scratch before moving on to MDAC. - Ta bu shi da yu 09:06, 20 July 2005 (UTC)
- Support. Well written, very readable (espesially considering the subject matter). Only (minor) thing that ought to get dealt with is the red links. WegianWarrior 09:20, 20 July 2005 (UTC)
- Hmmm... I'll be getting to it :) Thanks Wegian! - Ta bu shi da yu 10:39, 20 July 2005 (UTC)
- Support --Phroziac (talk) 13:40, 20 July 2005 (UTC)
- Neutral. Really an "object", but I'll be away the next few weeks, and I don't want to stand in the article's way because of some obsolete old objection that I couldn't strike out due to my absence. I trust Ta bu to address my points anyway. While this is good work, there are many open questions: Lupo 08:32, July 21, 2005 (UTC)
- I'm unsure whether the longish and yet superficial explanations of "Locking", "Transaction processing", and "Data integrity" really belong into this article. Wouldn't it be better to have full-blown articles on these subjects and just give summaries (with "Main article"-links) here? After all, it's not as if Jet was in any way special: these mechanisms are/were commonplace. Lupo 08:32, July 21, 2005 (UTC)
- Hmmmm... good point. I have created Lock (database). There are some specific things in that section to do with Jet, though. Transaction processing already exists, and I have Jet specific stuff in that section, so don't feel that a {{seemain}} would be appropriate for this section. I quite like the "Data Integrity" section, and feel that it is necessary background information. It's possible that it should be placed into its own article. Not sure, what do others think? The queries stuff is very much specific to Jet, same with security. - Ta bu shi da yu 04:13, 22 July 2005 (UTC)
- I'm worried about the accuracy of these descriptions. The article states, for instance, that "With pessimistic locking it is guaranteed that the record will be updated." This must be qualified by "if the user obtained the lock", otherwise the obvious question a reader will ask is "then why use optimistic locking at all?". The point is that you may get more lock conflicts with a pessimistic policy, which is why optimistic schemes were invented based on the observation that most transactions won't conflict anyway. Lupo 08:32, July 21, 2005 (UTC)
- Hmmm... my understanding might be faulty here. - Ta bu shi da yu 04:13, 22 July 2005 (UTC)
- OK, I see what you are saying. I have updated the article. - Ta bu shi da yu 07:38, 2 August 2005 (UTC)
- Hmmm... my understanding might be faulty here. - Ta bu shi da yu 04:13, 22 July 2005 (UTC)
- What locking policy (policies) did Jet employ? 2PL? See also concurrency control. Lupo 08:32, July 21, 2005 (UTC)
- Gulp... don't know! OK, definitely something that needs sorting out. - Ta bu shi da yu 04:13, 22 July 2005 (UTC)
- I presume Jet has read and write locks, but the article talks about locking only in the context of updating. Lupo 08:32, July 21, 2005 (UTC)
- Another fair point. - Ta bu shi da yu 04:13, 22 July 2005 (UTC)
- "Until the transaction is committed, the only changes that are made are in memory and not actually done on disk." Yeah, really? What about logging and/or file caches? Was caching left entirely to the OS? Strikes me as unlikely, but what do I know what Microsoft did... Maybe safer to rephrase to "Until a transaction is successfully committed, changes are only recorded in temporary storage and not yet in the database itself." or some such. Lupo 08:32, July 21, 2005 (UTC)
- I got that from a Microsoft article. Will dig it out and add a footnote. - Ta bu shi da yu 04:13, 22 July 2005 (UTC)
- Got this from Microsoft themselves:
- "Since the operations in a transaction are saved up in memory until the entire transaction is committed, application developers can benefit from their use even when a transaction would not otherwise be necessary" [1]
- The article fails to explain what a "user" is. I take it that it is a software that uses the Jet DLLs, hence if one human user runs two different programs that both use Jet to access the same database, these two programs would be two "users" in the context of this article. Is that the intended meaning? If so, explain it; a layperson might think the word refers to a human user. If not, also explain it! Lupo 08:32, July 21, 2005 (UTC)
- OK, will do. This is good stuff! - Ta bu shi da yu 04:13, 22 July 2005 (UTC)
- "Queries" talks about SQL queries, but the lead-in paragraph of "Architecture" states that the ability to run SQL queries was only added in later versions... Lupo 08:32, July 21, 2005 (UTC)
- Good point. - Ta bu shi da yu 04:13, 22 July 2005 (UTC)
- The lead paragraph states that Jet has been obsoleted by Microsoft SQL Server. When? When did MS stop selling applications based on Jet? Lupo 08:32, July 21, 2005 (UTC)
- Will check the MDAC article - it says it in there somewhere. I know it is definitely not being produced any more. - Ta bu shi da yu 04:13, 22 July 2005 (UTC)
- Is it "jet" or "JET"? The lead paragraph has both... Lupo 08:32, July 21, 2005 (UTC)
- Good point. - Ta bu shi da yu 04:13, 22 July 2005 (UTC)
- Now fixed. - Ta bu shi da yu 07:26, 2 August 2005 (UTC)
- Some minor grammatical/stylistic things such as the use of "till" instead of "until" (I'm a foreign speaker, so I may be mistaken, but "until" strikes me as more formal and more appropriate for an encyclopedia article; I've changed that myself) or "a Jet dynamic link library (DLL) that could directly manipulate Microsoft Access database files (MDB), which was a modified form of an Indexed Sequential Access Method (ISAM) database", which doesn't parse well because the singular last clause (on ISAM) refers to the plural mention of MDB, not to the singular "Jet DLL". (I haven't come up with a good idea how to rephrase this.) Another problem is that some paragraphs are written in the present tense and others in past tense. Lupo 08:32, July 21, 2005 (UTC)
- Ya. Agree it needs a good copyedit. - Ta bu shi da yu 04:13, 22 July 2005 (UTC)
- I'm unsure whether the longish and yet superficial explanations of "Locking", "Transaction processing", and "Data integrity" really belong into this article. Wouldn't it be better to have full-blown articles on these subjects and just give summaries (with "Main article"-links) here? After all, it's not as if Jet was in any way special: these mechanisms are/were commonplace. Lupo 08:32, July 21, 2005 (UTC)
- Strong support, especially like the table in the History section. Phoenix2 03:11, July 24, 2005 (UTC)
- Object I read the intro and couldn't work out what was going on. Needs to be restyled so that a layman can understand what it is (by all means have a few technical details later, but at least allow a layman to get an idea of what a Microsoft Jet Database Engine is, why it's important and what it does, jguk 07:37, 30 July 2005 (UTC)
Neutral, the table and graphic down the bottom (in the "History" section) are way to wide... Alphax τεχ 10:58, 31 July 2005 (UTC)- Changed to 300px. - Ta bu shi da yu 05:02, 3 August 2005 (UTC)
- Changing vote to support, much betterer now! Very detailed article on a difficult subject. Alphax τεχ 15:30, 3 August 2005 (UTC)
- Changed to 300px. - Ta bu shi da yu 05:02, 3 August 2005 (UTC)