User talk:Warren/0810

Development of Windows XP

In response to the IP editor's repeated gross incivility and refusal to engage on the talk page I have created a Wikiquette Alert. I wish we could go straight to a permanent block request on the IP, but this is the first step in the process. Jeh (talk) 02:14, 6 September 2008 (UTC)

Vandalism problem on the MASM32 Page

Warren,

Apologies for having to bother you but the page for the MASM32 project has been locked while containing illegal content. I am the webmaster of www.masm32.com and I am the copyright holder of the web site's content yet after repeated removing vandalism from a user in Denmark, the anon user has continued to post parts of the masm32 Project licence which I have never granted permission to any person or organisation to do.

The repeated re-edit and reversions performed by the anon user also contain polemic from the Usenet alt.lang.asm which includes a legal opinion by the anon user that cannot be verified. As I can most probably identify the person and I am familiar with his attacks and criticism of the MASM32 Project for the last 8 years or so, this matter will not be resolved by consensus.

Would you please contact me at my email address hutch@movsd.com so I can pass to you any information relevant to the problem generated by the anon user from Denmark.

I would like to be able to update the page to keep it current and in good faith with the published intent of Wikipedia but it is impossible to do while the page is repeatedly being vandalised.

Steve Hutchesson for the MASM32 Project.

Hutch48 (talk) 05:40, 11 September 2008 (UTC)

Hi, Steve. To be honest, I'm not really all that interested in who the anon you're talking about is. What matters (both to me personally, the encyclopedia at large) is that these sorts of content disputes do not take place on our article pages. This is really important. You can make use of Talk:MASM32 to discuss changes that need to be made to the article while it is protected from editing. Understand also that, by identifying yourself as someone with a vested interest in the subject, you may find yourself on the wrong side of Wikipedia's conflict of interest policy. We're here to build an encyclopedia, not to serve as anyone's marketing and promotion platform. I'm definitely not saying you've done anything particularly wrong here, nor you should feel discouraged from editing, but do keep this policy in mind when working on the encyclopedia.

Also keep in mind that Wikipedia does have a three-revert rule; editors who engage in repeated reversions of another person's work may be subjected to temporary blocks. If you are dealing with a troublesome editor who engages in such edit-warring, report them for a 3RR violation. Nobody's going to call you on this for simply removing your own copyrighted material from the encyclopedia; but do be careful about removing content that is critical of you or your software. Doing so will get you blocked from editing, and that's probably not what you're hoping to accomplish here. We have a really good page on Wikipedia:Dispute resolution which will give you lots of guidance on where to go from here.

One final thing -- generally speaking, Wikipedia articles on software should not contain sections on, or discussions of "Alternatives". We wouldn't do this with articles on albums, sports teams, schools, religions, philosophies, etc., right? Software articles need to adhere to the same standard. Always feel free to remove such sections in favour of creating list articles and categories that contain lists of related articles.

Good luck... Warren -talk- 18:25, 11 September 2008 (UTC)

Hi Warren, thanks for responding, the admin who originally locked the topic to protect its content has removed the illegal content and left it locked for the moment which is fine by me. The specific project has been running for 10 years as a freeware programming system that is not an item of commerce, it cannot be either bought or sold. The page was originally written some years ago by users of the MASM32 project and I note from the page history that a number of changes that were hostile were removed long before I noticed that the page existed.

It has been my experience that Wikipedia contains some very high quality technical data in the programing field and to this extent I would hope that such content will be safe from deliberate tampering by interested parties for whatever reason. The page for the MASM32 Project does need some modernisation as it is an adapting technical project but repeated deliberate tampering with the content to change it from a technical resource in Wikipedia to a vehicle for polemic is neither reasonable nor as best as I can determine consistent with the policies of constructing an online encyclopedia.

I know that Wikipedia has recently had to tighten up its policy on anonymous editing after unfavourable publicity in US based media channels and perhaps this is a good enough reason to enforce editing from registered and identifiable parties only as it prevents the page hacking that Wikipedia is currently vulnerable to. There is no reason why anonymous editing should be allowed when it is wide open to abuse of the type that defaces web sites through various exploits and weaknesses in internet languages and server side software.

As far as the policies of Wikipedia, tolerate me here for a little while, the learning curve for a text mode interface of near infinite nesting of page after page complete with automated resolution pages makes responding to vandalism a daunting task for anyone who is not highly familiar with the Wikipedia interface. The weakness with the 3 reversion rule is it leaves the page defaced while waiting for someone to respond and while I am happy enough to work within the rules of Wikipedia, the page hacker does not have to bother and with the ease of changing and IP to avoid detection, the system favours the hacker.

AS far as editing the page, I am more than happy for competent technical data to be added to the page. The MASM32 Project is older than Wikipedia as has long had the contributions of many very skilled technical people added to it. My problem is not with competent technical data but deliberate hacking and using the Wikipedia inernet capacity as a vehicle for extending polemic that originated in other venues.

Regards and thanks for your contribution to protecting the page from vandalism.

Steve Hutchesson for the MASM2 Project.

Hutch48 (talk) 02:28, 12 September 2008 (UTC)

AfD nomination of Nikhil Kothari

An article that you have been involved in editing, Nikhil Kothari, has been listed for deletion. If you are interested in the deletion discussion, please participate by adding your comments at Wikipedia:Articles for deletion/Nikhil Kothari. Thank you. Do you want to opt out of receiving this notice? I do not know Nikhil Kothari, but it appeared as a pattern in terms of same IP updating the article over a period of one year. It is also that the article may not have much significance. Chirag (talk) 19:50, 11 September 2008 (UTC)

Merger of WikiProject Free Software and WikiProject Software

Greetings, I have made a proposal for the merger of WikiProject Free Software and WikiProject Software here. I invite you for your valuable comments in the discussion. You are receiving this note as you are a member of the project. -- Tyw7, Leading Innovations ‍ ‍‍ (Talk ● Contributions) 20:02, 13 September 2008 (UTC)

Post-dated fact tag

I finally looked up that "post-dated" fact tag that you apprised me of. All I can say is that I must have been tired and probably assumed that the material had been around since November of 2006 because that was the date given in the article for the Media Research Center's press release. As for your point about providing documentation when it's missing, there are two ways of looking at it. In the short run you are improving articles. In the long run you may be harming the whole enterprise by encouraging others to be lazy. If you have kids you can appreciate that. Cheers. Badmintonhist (talk) 04:36, 16 September 2008 (UTC)

Speaking of providing documentation Warren, ol' boy, a brief search through the article on MSNBC or in Google would have yielded the missing sources in the MSNBC intro. But now maybe you believe in letting the people who included the information document it. Badmintonhist (talk) 22:38, 3 October 2008 (UTC)

Because it is clearly his main gig

Ah, Warren. Warren, Warren. You did catch me on my failure to capitalize "ideology." As for the description of Countdown in the "Keith Olbermann" article's introduction, surely the show, as Olbermann 's "signature" claim to fame (and the whole reason for this lengthy Wikipedia bio) merits a slightly more detailed description than does his role as a football pre-game host. How much of the rest of the article deals with his football analyst role? 'Tis a small point and one that I don't plan to pursue (right now). However, you might use your usually impressive ability to think logically and to put things in perspective to improve this and other politically "sensitive" articles in more substantial ways. Badmintonhist (talk) 23:46, 21 September 2008 (UTC)

D&D articles for Wikipedia 0.7

Hi there!  :)

As someone who's worked on D&D and/or RPG articles before, I'm inviting you to participate in our goal to both improve articles that have been selected to be placed in the next Wikipedia DVD release, as well as nominate more to be selected for this project. Please see the WikiProject D&D talk page for more details. :) BOZ (talk) 19:36, 25 September 2008 (UTC)

FAR: Virtuti Militari

Virtuti Militari has been nominated for a featured article review. Articles are typically reviewed for two weeks. Please leave your comments and help us to return the article to featured quality. If concerns are not addressed during the review period, articles are moved onto the Featured Article Removal Candidates list for a further period, where editors may declare "Keep" or "Remove" the article from featured status. The instructions for the review process are here. Reviewers' concerns are here. --ROGER DAVIES ^talk 15:16, 3 October 2008 (UTC)

Weasel and lead

Warren, Fox News Channel had this very discussion several times over the course of a few years, and the wide consensus has been that inserting individual critics and/or references in the lead gives them undue weight. Weasel words is a guideline, and there used to be (if not still) a specific caveat in WP:WEASEL that makes exceptions for weasel words when sources are "sufficiently numerous." Since verification occurs via the section in the article, all is well. If those who argued for inclusion on MSNBC based on the fact that Fox News Channel mentioned bias controversy in the lead, they should also understand reasoning behind it (confirmed now by 4 RFCs). See Talk:Fox News Channel/FAQ. Thanks! //Blaxthos ( t / c ) 01:58, 4 October 2008 (UTC)

I see you're busy butchering the Conventional Memory article

The issue of modern memory 32-bit limitations vs the historical conventional memory limit is a discussion of the fact that these system architecture limitations have not gone away and still remain with us, and likely will continue.

Also the section about how 640k would have been "considered enough" that you're busy axing is an issue of historical perspective, due to the fact that DOS itself at the time fit in less than 50kb and most programs weren't any larger, and the very first IBM PC motherboards could only be equipped with 64k to 256k at maximum without an ISA memory expander board. If IBM itself limited the motherboard to 256kb at maximum that in itself is a clear indication that they thought it was considered enough memory for the first release of the product.

I think most of your deletions should be reverted as being too narrow-minded about the topic being discussed. DMahalko (talk) 21:38, 6 October 2008 (UTC)

Couldn't possibly be because you wrote that section, could it?

Look. The term "Conventional memory" refers specifically and only to the first 640kb of real mode x86 on the IBM PC architecture. It's not a term you are entitled to stretch to mean something that it doesn't. Nor is it suitable to This isn't a matter of being narrow-minded; this is a matter of writing the correct information into the correct articles.

If you want to write on the subject of the 4 GB addressable memory limitation on what is actually called the PCI hole by Sun, by all means, click here and go for it.... but do it in a new article. I don't really have any time or interest in writing that article right now... perhaps you do.

But before you go ahead with that, I have to remind you that the "3.2 gigabyte barrier" section you wrote in the conventional memory article was deleted outright because you failed to provide any reliable sources at all. You actually made the term up! Lack of attribution is a big problem on Wikipedia; people write stuff into the encyclopedia based on their own experiences, or maybe stuff they read about on forums or heard from friends, but don't provide proper, reliable sources as is required by Wikipedia's verifiability policy. Surely you aren't expecting that you can write just anything off the top of your head, into the encyclopedia, and not have it challenged or deleted at some point. Attribution ensures that the information we publish is accurate -- or, at least, the inaccuracies can be be faulted on our sources. Warren -talk- 23:43, 6 October 2008 (UTC)

Ah, so you're of the camp that says that if the editing isn't just a stellar scholarly effort that cites every single paragraph in the very first version, it must be instantly deleted as unacceptable.

I see article editing as a collaborative effort where people who may not fully understand a subject can write something which is close to being accurate but may not be fully correct due to their own knowledge limits, but which someone else more knowledgable can come along later and correct.

The 32-bit memory hole issue is poorly understood by most people (including me, which I directly admit) but there has nothing else better available than what I've written about it on Wikipedia. So while it's certainly not as accurate as you think it should be, it is (was) pretty close. All it needed was a few corrections from someone more knowledgeable of the subject such as yourself, except you cannot be bothered to take the time to fix it or move it to a more appropriate subject heading, and instead you're going to summarily kill it on the spot without further thought. This to me is destroying good faith edits, which is where you're breaking Wikipedia policies as well.

The Otto engine article I recently started is not as absolutely perfectly perfect as you may perfectly want it, and it will likely never have any good cites because they are simply not available. However the content is about as good as it gets. I suppose you're going to pop over and delete all that now too since I failed to cite any reliable sources. DMahalko (talk) 02:02, 7 October 2008 (UTC)

Sorry but no, your discussino of the "32-bit memory hole issue" was not "pretty close" at all. For example, 64-bit hardware is not required to work around it, only full support of Physical Address Extension. And as Warren said this topic is completely irrelevant to "conventional memory". Jeh (talk) 05:36, 7 October 2008 (UTC)

p.s. here is a good description of the issue. Note that the issue is specific to "consumer" versions of Windows. Windows Server 2003 (32 bit) doesn't have the issue. Nor do various versions of Linux. Jeh (talk) 05:39, 7 October 2008 (UTC)

Conventional memory was an arbitrary structural design limitation due to some address space being reserved for hardware use. The (variable) loss of memory for (some) 32-bit operating systems is due to the same reserved-hardware-region issues, as documented on that very Russinovich page your "citing at me", and thus the past limitations of the conventional memory architecture repeat themselves in modern architectures -- which is the whole point that article section was trying to make, and which is the point you are missing completely. Providing a historical perspective of how past design limitations have repeated themselves in subsequent "modern" system architectures does not distract from the primary topic at hand. DMahalko (talk) 06:29, 7 October 2008 (UTC)

Perhaps not in a magazine article, but it does distract in an encyclopedia. There's a difference in styles. Jeh (talk) 07:14, 7 October 2008 (UTC)

Ah, but now we get to the core of the issue, where you are trying to justify not including anything at all and justifying Warren's complete removal of my good faith edits. He couldn't defend this action against wikipedia policy himself so he's brought in a "hired gun" to debate it for him.

That is a completely false accusation against Warren. He didn't "bring in" anybody. I happen to have Warren's talk page in my watchlist because we've had other exchanges recently. I'm here purely on my own volition. Jeh (talk) 19:17, 7 October 2008 (UTC)

PAE which you are citing as a 32-bit workaround, was originally only something that special server software utilized, such as high-end database programs. The fact that more and more modern Windows programs can support PAE does not obviate the fact that the 32-bit hardware limitations affected (and still affect) many computer systems, just as conventional memory forced people to fool with DOS extenders and drivers that had to be built to "load high" way back when. Oh, but I guess these analogies to the past are getting lost on you. DMahalko (talk) 06:29, 7 October 2008 (UTC)

The point of bringing up PAE to you is to show you that you are mistaken when you claim that 64-bit hardware is required to get past the 3.2 GiB "barrier". Support for a full 4 GiB, or even more, of RAM has nothing to do with 64-bit hardware. It has to do with how virtual addresses (which are the addresses that are 32-bit in a 32-bit environment, and 64-bit in etc.) are translated to physical addresses (whose bit sizes come from the page table entries). PAE is not something that "programs" can get to at all in a protected OS environment, and it is not a "workaround" (despite what the white paper from Sun Microsystems says). It is the same address translation scheme that has always been done, with one more level of table lookup and wider page table entries, the latter providing more bits of physical address than are possible without it. It's something the operating system supports and (again, despite what the Sun white paper says) is completely transparent to application programs, since apps use only virtual addresses. By supporting PAE, an operating system can not only use all 4 GiB of RAM in a 32-bit system so equipped, it can use considerably more. How much more depends on the specific chip, but the first CPUs that supported PAE could generate 36-bit physical addresses, hence 64 GiB of RAM. Incidently on x64 in "long mode" the page table format looks very much like PAE-format page tables, but with yet an additional stage of lookup table. Jeh (talk) 07:14, 7 October 2008 (UTC)

For all this technical blather, it does nothing to resolve the fact that such architecture limitations exist and continue to plague modern computer systems. The very computer I write this with, outfitted with dual 512meg 8800GT video cards, 4 gig of memory, and Windows XP 32-bit (hardly an obscure operating system choice) shows only 2.75 gig available. As such this is a perfect example of such architecture limitations. The fact that you are arguing around it with very specific exceptions and with very specific attacks on what I wrote (3.2 vs a variable size) does not invalidate the fact that this (still) is a very real problem. DMahalko (talk) 18:28, 7 October 2008 (UTC)

I am sorry I didn't explain myself clearly, for it appears that you are missing the essential point: What you are calling the "3.2 GiB barrier" is not at all analagous or historically tied to or related to the 640 KiB barrier, because the 3.2 GiB barrier is not an architectural limitation of either the hardware or the software, or the combination of the two, as was the 640 KiB size of "conventional memory". Jeh (talk) 19:17, 7 October 2008 (UTC)

You're probably going to call me on the fact that I didn't point this out before, and it's true, I didn't; I hadn't thought it through. But now I have, and this is the conclusion. The "3.2 GiB barrier" is merely a choice made by the developers of Windows "desktop" operating systems, XP SP2 and later. These operating systems are architecturally capable of letting you see all of your 4 GiB RAM on your system, but there is a purely artificial limit in them, such that they will not access RAM with addresses higher than 4 GiB. Were you running any of several other operating systems (including some 32 bit ones, and even some Windows 32-bit ones) you would be seeing all of your 4 GiB RAM despite the "PCI hole." Jeh (talk) 19:17, 7 October 2008 (UTC)

"Technical blather" is apparently uninteresting to you, but even so, I will point out that XP SP2 does enable PAE for purposes of enabling the no execute feature. i.e. unless you're either a) running on very old hardware (which you're not if you're running those video cards), or b) running with noexecute disabled completely, you're already running with PAE turned on. But apparently some third party driver writers for some common consumer hardware just can't be bothered to support physical addresses wider than 32 bits, so (after a large number of online crash analysis reports) the Windows kernel team decided to artificially disable access to physical addresses above 4 GiB on consumer OSs, XP SP2 and later. Jeh (talk) 19:17, 7 October 2008 (UTC)

As I see it, about the only valid relationship that can be drawn between these two "barriers" is to note that despite superficial similarities they do not have common origins and are not really related to one another. Jeh (talk) 19:17, 7 October 2008 (UTC)

Your primary intent here is to just throw in my face how wrong wrong wrong I was to write anything at all, but as I say nothing better was available, and now nothing better IS available at all due to Warren's complete deletion of that text. Clearly you should be writing this PCI hole article rather than me since you obviously understand the topic so well, but it appears to me your primary interest here is just to show off your superior technical knowledge over me rather than trying to actually contribute to the encyclopedia, which is what the original intent of my editing was attempting to do. (Spirit vs Letter) DMahalko (talk) 18:28, 7 October 2008 (UTC)

Hmm, nice lack of AGF on your part... No, my primary intent here is to dissuade you from attempting to write further about a subject which you yourself have stated you understand only poorly; it will only cause knowledgeable people to have to do more work to clean things up after you. I realize that when you have written about something it is difficult to accept deletion. But really, the only factually correct alternative would have been to invert it so that it said pretty much the opposite of what you were claiming: "Despite appearances, the limit of around 3 GB RAM seen on some Windows operating systems is not related to the 640 KB "conventional memory" limit, as it is not an architectural limit of the hardware or the software." And if you put your text back, that's pretty much what I'll have to do to it. Jeh (talk) 19:17, 7 October 2008 (UTC)

That "nothing better was available" does not mean that any old text was acceptable, despite its making many erroneous claims; nor is it necessary to retain "that which is not so" because of AGF. The fact that "that which is not so" was written with good intentions doesn't mean that it, or any edited form, has to be or should be retained. Jeh (talk) 19:17, 7 October 2008 (UTC)

It appears you (and Warren) just don't get it so I am done with this discussion, but I'm not done with that article. DMahalko (talk) 18:28, 7 October 2008 (UTC)

I'm not done with correcting, or deleting completely, that-which-is-not-so either. Jeh (talk) 19:17, 7 October 2008 (UTC)

If you're going to wave linux as an excuse to not discuss the modern 32-bit memory limitations, note that various early versions of linux did not suffer from the conventional memory limit either (oh dear another analogy, I realize you hate these) so we can probably delete this whole article in that case. DMahalko (talk) 06:29, 7 October 2008 (UTC)

No, I "waved Linux" to show that the "barrier" you wrote about is not universal to 32-bit operating systems and so does not require 64-bit hardware to get past. Again you attempt to counter arguments that were not made in the first place. Again, as Warren suggested, if you want to write about the PCI hole, please do so; then you could put a link to that article as a "see also" under "conventional memory". Jeh (talk) 07:14, 7 October 2008 (UTC)

I'm going to de-indent here and address a few issues brought up in the above discussion:

1. I'm in the "Write Wikipedia as the policies demand" camp, DMahalko. You, apparently, aren't. I'm not interested in your accusatory tone on this point -- if you refuse to, or unable to provide references for anything you write in the encyclopedia, it can be, and eventually will be removed. That's how Wikipedia works -- that's how we ensure the content of the encyclopedia isn't stuff that people made up. If you don't like that, find another project to work on, because Wikipedia isn't the place for you.
2. Yes, DMahalko, having no information at all is better than having poor, misleading, mislabeled, unsourced, and essentially incorrect information. We can't do any harm if we don't have any information on a topic, but we will do harm if we have bad information. Surely on some level you must agree with the philosophy that you'd rather have someone tell you "I don't know", than lie to you.
3. The PCI hole issue is not, strictly speaking, an issue that's limited to consumer editions of Windows. Russinovich's article notes that the problem is related to drivers that don't know how to do address remapping. This has been a fairly common bug for driver writers for Windows since the beginning, but it's gotten a lot better in recent years now that it's common for people to have machines with 4+GB of physical memory. It's important to not confuse license limits (which are indeed arbitrary) with effective limits imposed on the operating system by hardware & device drivers.
4. I realise that it's fashionable in Internet debating to go off on snarky tangents like "well Linux didn't have the conventional memory problem, so we should just delete the whole article on conventional memory!" ... this doesn't help the discussion. Linux does actually have to deal with the same PCI hole issue as every other operating system, but because there is much better control over driver quality, and because there are far fewer drivers, the issue doesn't manifest itself as often, and Linux didn't need to do what Windows had to do in terms of artificially limiting 32-bit operating systems to 4GB physical memory.

However, there are a variety of older motherboards that are buggy and cannot do proper remapping of physical RAM above the 4GB point so that the operating system (doesn't matter which can access it instead of having it hidden behind PCI devices. I found a discussion of the problem affecting Linux 2.6.25 here, and there's lots more to be found on the Internet about it.

I'd like to think that some focus could be put on writing a proper article on the topic, instead of arguing over whether this information belongs in an article about a related, but definitively different technical problem. Warren -talk- 01:19, 8 October 2008 (UTC)

It's almost over...

Wikipedia:Suspected sock puppets/Fantasy Game Productions - Josh (talk | contribs) 04:51, 28 October 2008 (UTC)