jaclaz

That's good , I will gladly agree to a "softer "definition, what do you call something that can be proven wrong by reality or that another/study/poll/statistical analysis can contradict? However, statistics has some diignity, , ESTIMATED google counts do not (and they are NOT "statistics"). jaclaz

Sure it can be done, see the first post on this thread, as long as the recovery .iso can access the internal hard disk, the same command would work alright. But I am missing the need/usefulness of a temp file. jaclaz

Exactly, you can measure popularity by making a poll, then, provided that your sample is large enough and casual enough you can, using statistical formulas (which may include contrasting theories and consequently different results) extend your (limited) sample to the totality. It is in any case an educated guess, which does not demonstrate anything. You are expressing (mathematically and statistically) a synecdoche: http://en.wikipedia.org/wiki/Synecdoche (the part for the whole) So at least you have a "real" (though limited) base of data. (from which you can anyway draw the wrong conclusions) In the case of google results counts you have NOT this base as the google counts are ESTIMATES: https://developers.google.com/search-appliance/documentation/62/xml_reference?hl=it-IT#appendix_num_results now, if JorgeA would manually examine (or write an "intelligent" program doing the same) each and every page returned (not "counted" )by the google search, dicarding duplicates, false results, etc. then he would have a "base", i.e. a verifiable list of web pages where the intention of boycotting Windows 8 is actually expressed by someone. Fresh google results: jorgeA = About 5,970,000 results (0.23 seconds) HalloweenDocument12=6 results (0.26 seconds) jorgeA is BY FAR more popular than you . jaclaz

Demonstrating is impossible. Attempts to it are futile. We have all before us examples of how well paid specialists in "market reserch" or "trend forecasting" often publish (basing themselves on far more "solid" data than a lousy google search results ESTIMATE) completely wrong conclusions. Or look how "exit polls"(which are much more "tangible" and subject to statistical corrections and what not) often, when the result of the votes becomes official can be often largely disproved. No it makes it a statement, NOT a demonstration. This statement, which is however a guess, educated as it might be, remains a statement, and nothing more Popularity exists, the way it is measured is fallible, and of all the methods using the ESTIMATION of google results is one of the more fallible, the given link: http://homepage.ntlworld.com./jonathan.deboynepollard/FGA/google-result-counts-are-a-meaningless-metric.html explains in more detail why exactly. jaclaz

There are NO methods that offer *any* kind of reliability, and certainly not for a common word like "windows". BTW I just showed you how small diferences (such as the "broken windows 8" example) can lead to VERY different results. Fresh from my google: boycott Linux = About 870,000 results (0.24 seconds) boycott Mac OS = About 386,000 results (0.23 seconds) boycott "Mac OS" = About 333,000 results (0.25 seconds) boycott Microsoft = About 3,010,000 results (0.33 seconds) boycott windows = About 3,700,000 results (0.34 seconds) boycott windows 8 = About 3,000,000 results (0.31 seconds) boycott "windows 8" = About 425,000 results (0.30 seconds) boycott iOS =About 519,000 results (0.28 seconds) So, we have a similar number of people talking of boycotting "Windows 8" and Mac Os, and almost double talking of boycotting Linux. Moreover it seems like the people talking of boycotting Microsoft or Windows in general are many more than those talking of boycotting specifically "windows 8", which are about the same amount of those talking of boycotting Mac Os or iOS, that means that Windows 8 is a success . Meaningless metrics, more: boycott surface = About 3,350,000 results (0.32 seconds) boycott surface RT = About 40,400,000 results (0.60 seconds) boycott "surface RT" = About 210,000 results (0.33 seconds) boycott iPad = About 9,760,000 results (0.30 seconds) boycott Apple = About 4,970,000 results (0.30 seconds) Interestingly: boycott Asus = About 2,860,000 results (0.33 seconds) boycott Acer = About 3,060,000 results (0.38 seconds) boycott Samsung = About 2,180,000 results (0.30 seconds) boycott google = About 23,900,000 results (0.27 seconds) Samsung make better products than both Acer and Asus and is far more popular than Apple (and incredibly MUCH less adversed than google). jaclaz

If you want to lower yourself at the level of those we tend to criticize in this thread, go ahead. Before that, please try googling for: YES, with "windows 8" enclosed in double quotes. Another example: broken windows = About 72,000,000 results (0.33 seconds) broken windows 8 = About 151,000,000 results (0.26 seconds) broken "windows 8" = About 14,400,000 results (0.26 seconds) If you want to insist on this meaningless approach, you are obviously free to do so. The word for today is polysemy: https://en.wikipedia.org/wiki/Polysemy jaclaz

TurboPascal 5.5: http://web.archive.org/web/20060322203829/http://bdn.borland.com/article/0,1410,20803,00.html A few more "ancient" versions: http://web.archive.org/web/20060321230133/http://bdn.borland.com/museum Lazarus: http://www.lazarus.freepascal.org/ jaclaz

The shutdown is logos.sys and the wait is logow.sys Add it to C:\Windows, i.e. add a file C:\Windows\logos.sys, this should apply: http://www.themeworld.com/support/walls-change.html jaclaz

Let's see THREE questions: TWO answers: None of them descriptive, let alone "actual", "exact", "complete", and "precisely". NO, you are not providing enough meaningful data, of course if you want some botched guess, that's OK, examples: jaclaz

Look, the general idea is to solve issue (if possible), not to solve them AND solve them in the way you thing "nice" or "easy" or simply "fancy". This may be "hard" , but works: Dukem.cmd @eduke32.exe -nologo -D%~nx1 In file association, "open" command: "nircmd.exe" exec hide Dukem.cmd "%1" jaclaz

Then WHY are you asking if it is a bug of a software (which BTW you didn't specify initially). Please do provide the standard litany: http://homepage.ntlworld.com./jonathan.deboynepollard/FGA/problem-report-standard-litany.html Till now you provided not enough data to even guess what the issue may be. jaclaz

Yes/No. I just checked the file in the Czech 2K SP4 and it is NOT the same file (different size and checksum) though the patch, as guessed initially is confirmed to work (the offset remains the same 0x3887). It is definitely a "different" build, though thee differences are trifling at first sight. jaclaz

DO NOT use Google Search to support anything, siimply don't! http://homepage.ntlworld.com./jonathan.deboynepollard/FGA/google-result-counts-are-a-meaningless-metric.html IMHO it would be more practical to make a list of what WILL NOT be affected, I already made an EXHAUSTIVE list of these apps : Notepad Paint Solitaire This won't in any way affect the productivity of the average user. Just in case : Bart's chalkboard jaclaz

Yes, I got that, but we don't know if anything like this: actually happened (and remained "sticky") . Maybe there is a way to re-install the NTbackup and/or to reset correctly the permissions. Which specific permission error are you getting? jaclaz

Sure, there are Rules (and personally I tend to comply with them) the whole point being that you have NO way to know WHICH specific Rules is Brando569 subject to. If instead of the "scary tactic" implied in: you had used a plainer : I wouldn't have commented on it. jaclaz

I guess that there has been a misunderstanding. I have no idea if the patch to setupapi.dll that FDV posted about works on a czech system. As said FDV has tested the patch on the English version and I remember having used it on the Italian Win2K SP4 years ago. You cannot take for granted that the patch will work on *any* setupapi.dll, I was telling you four things: the patch was intended for En systems ONLY the patch has worked for lots of people FDV was not in any way "wrong" in connection with this the patching process is not a "challenge", and the instructions are seemingly very clear to follow but since the patch (though at different location within the files) is the same for several versions of the setupapi.dll, I am lead to believe that it should work also on "localized" versions, though you never know with anything "MS". It is perfectly possible that the patch does not work on Czech source, I have no way to know, if it doesn't work, you may *try* to patch an English setupapi.dll and replace the Czech version. AFAICR there is no need of the patched .dll to edit TXTSETUP.SIF. jaclaz

Winoutreach5, When looking at your replies I have to wonder if you are actually providing any specific solution to specific problems or only some general pushing towards using only MS tools and upgrading to latest MS Operating System versions, and the generic notion that MDT based deployments can fix *everything*. The case at hand is - seemingly - an issue originated by the "wrong" behaviour of some MS original updates, connected with "localized versions". Does MDT deployment "automagically" fix such problematic "localized version" updates? jaclaz

You can add one or more parameter to the file association command, see: http://www.la-solutions.co.uk/content/fileassociations.htm http://www.vbaccelerator.com/home/VB/Code/Libraries/Shell_Projects/Registering_File_Associations/article.asp If this command works for you in a command line: eduke32 -nologo -Ddemo8.dmo and you make an association to (say) file type .dmo, you will have in it something like: "C:\whatever_path\eduke32" -nologo -D"%1" or "C:\whatever_path\eduke32" -nologo -D%1 or "C:\whatever_path\eduke32" -nologo "-D%1" If it doesn't work, you can use an indirect way, by using cmd.exe (but you will have an open "cmd window") or Nirsoft nicmd.exe (that can be run in "hidden mode") and have them call the eduke32 as if it was done on command line. http://www.nirsoft.net/utils/nircmd.html jaclaz

No, you got it completely wrong. The "final" thing will be called "W" (easier to remember and takes only one byte space in RAM / one neuron in brain, more suited for the limited resources of the intended users) and will come in a single edition, one size fits all (this allows for reducing developers/branches) from crappy phone to top-notch enterprise servers, instead of having an original idea stolen from "The Matrix", they will use an original idea stolen this time from "Men in Black" (go to 0:54): The real issue being that there is a secret department in Redmond working on a neuralyzer app. (one that works, not one of the joke apps for Android and similar) They will finally be able to erase your memory (beside your hard disk and all the data you have on the cloud) remotely.... On the cited Thurrot's site, there is an interesting comment by user John Galt: http://winsupersite.com/windows-8/what-blue-leak-tells-us-about-microsoft-s-strategy do read the whole comment, it raises very good points, maybe it's "us", not "them". We will neededly all migrate eventually to Linux (or any other "working" OS) and let "them" fight senseless Surface vs.iPad wars. jaclaz

No , "delims" interprets consecutive separators as a single one. There is more than one way to skin a cat (the cat won't be happy anyway about it ), but the "real" issue as I see it is the comma inside the last field. You dont' need to use gsar for doing something *like*: FOR /F "tokens=*" %%A IN ('TYPE 1.txt') DO SET Line=%%A SET Line=%Line:,,=,"",% and then go on with processing "Line" instead of "1.txt", but the script you posted (at first sight) seem to me a rather complex one. If I get it right the idea should be to "ignore" the comma when it is inside a double quoted field.... I'll see which trick I can find to get that result in a simpler way (if I can find one) . jaclaz

No (not only "collisions"), which JFYI are normally automatically "solved" at boot time. (on XP/2003 and earlier, the article you cited is for Vista or later which use a slightly different boot approach), The Disk Signature affects Drive letter assignments. There are simple procedures to re-assign (within the OS on second disk) the SAME C:\ drive letter to the copy/clone. Changing the drive letter of an installed 2K/XP OS is doable, but it requires a number of steps/modifications that are so many and so complex that it simply makes attempting it foolish as you have NO guarantee whatsoever to have fixed "everything". As suspected : you managed to run that system in "mixed mode" with some (most of ) files used actually residing on "D:\" and some residing on "C:\". This does matter and can cause ANY kind of issues. Running a system in this "mixed mode" can additionally (and probably this is what happened/is happening) "corrupt" pemissions/links/Registry paths and what not, as the OS will try to "fix itself" some inconsistencies. Of course you won't notice anything particular for most "basic" programs/operations, but as soon as you use a complex tool (like NTbackup) you will probably (actually it is what you just reported) hit a brick wall. jaclaz

... and all these years I thought that DIR /X was made for that ... jaclaz

Capsbuster, with all due respect , it's the second time today that you post about something (that has worked for hundred or thousands or more people) being "wrong" (because *somehow* it doesn't work for you) You might want to be more cautious and check and re-check everything twice before saying other people or their contributions are "wrong". If you look attentively at this image: http://www.vorck.com/windows/edit-setupapi.html you may notice that it has superimposed TWO RED ARROWS, one highlighting the hex string and the second (left bottom) highlighting the address (3887). No actual challenge: Of course any info you will find of FDV's site is tested on English source files, BUT if you found the same bytes at the same location the SETUPAPI.DL_/SETUPAPI.DLL is most probably exactly the same file in the English and Czech versions of 2K SP4 (and, as a mere anecdotal reference, the hexediting did work at the time on my Italian version) jaclaz

Well, it's your files, but it seems to me a "queer" way to manage a project (or versioning). Basically you have not on that page the "stable" 0.2.3 (which would be the "logical" thing to use if you come here the first time), you have the 1.0beta8 (which, comeon , is Beta since more than one year, has contributed to install more XP's than you might ever think ), but you go around: suggesting the beta9 (which is experimental BUT not listed by default). IMHO a project should have: last "previous" versions (just in case) <- let's say 0.2.3 <-Marked as "Previous stable" "current" version <- the 1.0Beta8, aptly renamed to 1.0 <- Marked as "current" "experimental" version <- the 1.0Beta9, aptly renamed to 1.0Beta <- Marked as "experimental" these three versions should be available on the "default list", while all the versions but the three above should be not listed by default, but available in the advanced list: "previous" versions (just in case) <- let's say anything before 0.2.3 and all the 1.0beta's <-Marked as "Deprecated" jaclaz

Sure, and depending on what sites you access from the non firewall filtered connection you may have the g-men break into your house at 4:00 AM , taking you to Gitmo, without the possibility of any legal assistance. Come on . jaclaz