pointertovoid

Thank you! I'll definitely have a try - and even several ones. I'm still interested about hints and instructions to install KernelEx or KDV for FreeCad only, that installs on W2k but needs Xp dll entries.

"Cryptography is one thing you can rely on", said Edward Snowden. That's why governmental agencies, as they can't decipher the contents, prefer to tap the information elsewhere: at the targeted computer where the message is typed or read, or at the service provider who should introduce backdoors in the encryption - despite this weakens the own country's security against foreign attacks too. Presently, the French police minister tries to move his European fellows to let some Internet services providers, like the encrypted message exchange provider Telegram, to spy on their users and provide the data to the police. http://www.lemonde.fr/police-justice/article/2016/08/24/les-messageries-cryptees-dans-la-ligne-de-mire-des-responsables-politiques_4987058_1653578.html The excuse is that the murderers who slaughtered a priest in France last July had used Telegram. But this logic has flaws: A foreign intelligence service warned the French in advance of the priest's murder, so it's possible to get information despite Telegram, somehow. By being smarter than the French spooks maybe? The French had been warned and had even received names "but didn't have face pictures". The murderers had disappeared for weeks. They lived several days long under a tent in the centre of the target town. They kept unnoticed, kudos to the French. The terrorist of the Paris attacks in November synchronized themselves with an unciphered SMS. One mastermind could even pass the reinstalled border control after detailed verification. No relationship with cryptography. Most terrorist attacks are done by intelligence services, including the friends and allies, not to mention the FSB that can become a big threat if the Ukrainian or Syrian conflicts worsen, rather than by ISIL. They only avoid to go public about the attacks. These worse attacks are made easier by weaker privacy and secrecy in the own country or Union. But be sure that foreign secret services have better means than Telegram, and a legal move won't change that. How much is a people's freedom and privacy worth? The French's ancestors gave more than 300 lives for it. That's as many Frenches as tobacco kills every day. And shall the people ask for protection by the government, or protect themselves against the government? The French services spy on me and censor me, not the terrorists. The kind of weakness governmental agencies introduce is nothing benign nor unnoticed. Remember when the "bugged" Apache https servers chose their random key among only 65,536 instead of billions of billions (and the NSA bragged in an internal paper about having put a backdoor at the https): the very first thing cryptanalysts all over the world check is whether there are regular patterns among the keys, and here observing just 300 keys would have revealed that they just reappear - impossible to miss. Be certain that the FSB knew and exploited the backdoor, the only reason why they didn't tell the flaw. They could have broken down all banking activity and all bank accounts in a target country if they had wanted so.

Calling for more entries in other dll: yes, I got that from my Win98 time, and decided then not to do it again. I have a feeling that this would be stealing intellectual property (well, I have Xp licenses too) and also that MS has taken steps to hinder users in this attempt. That why I consider BWC's udpdate, yes. Freecad installs and runs on Xp, officially and experimentally, and to my understanding, BWC provides all XP dll entry points in his extensions, so detailing the used entry points shouldn't be necessary. Just a matter of what extended dll to install where and how, or at least I hope so.

Hello dear friends! I'd like to use FreeCad (a 3D CAD meant to be complete and free, presenly v0.15, beware 170MB installer plus 80MB manual) on a plain vanilla offline W2k. The installation goes without complaint but the application fails because Kernel32.dll doesn't offer the necessary entry points. This looks like a perfect use for KernelEx or KDW or some family member. But how should I proceed for this single application? Paste one or several improved dll in the application's folder? In System32? Or do I need deeper changes, which I'd prefer to avoid? Thank you!

GhostScript is needed by GSview. Both v9.05 and the latest v9.19, from 2012 and 2016, install and run properly since GSview runs. v9.05 is contemporary to the latest GSview v5.0, giving me more confidence. GSview is needed by some Latex editors to display Postscript .ps and .eps files. The latest v5.0 install and runs properly, or at least it displays nicely the files I have. GSview would also display .pdf files but is slower than Foxit and Sumatra. Some users report worries with GSview and Foxit as they lock the displayed file, preventing the Latex compiler to update it, and suggest Sumatra's behaviour fits this better. Postscript Viewer too displays postscript .ps files, and its latest v1.2 installs and runs on W2k. Though, it just converts the file to Pdf and calls the pdf viewer, and the conversion isn't fast: 1.5s for 350ko on an E8600. The nice news is that Sumatra displays Postscript files too. It's faster than GSview and Postscript Viewer, the display is as good, and Sumatra can search a string in the displayed document and jump to a page. Free, no splash window neither. Sumatra is known as a Pdf viewer. In v1+ it outpaces GSview and even Foxit a bit. It displays noisy jpg better than they do. v1.1 installs and runs on W2k without helping it. v1.51 is the latest that installs and runs if you add Gdiplus.dll in the installer's folder. My Gdiplus is v5.1.3102.1360, beware different versions meant for different applications seem uncompatible. Deinstalling needs to run uninstall.exe from the executable's folder (add/suppress programs doesn't work) with a Gdiplus.dll added there. v1.6 is the latest that installs and runs if adding Gdiplus.dll AND if refusing the Pdf search extension at install - or neglect the installer's error message. v1.7 fails to install and I didn't try to paste on W2k an installation made on Xp. Portable (zip) v3.12, v2.52, v1.9, v1.7 pasted in the executable folder fail to run. That is, the same that fail as an installer (exe). Foxit is a Pdf viewer that avoids Acrobat. v4.31 installs and runs, v5.4.5.0124 fails to install. Aspell comprises dictionaries (just lists of words for spell check) for many languages and a separate engine. Unsurprisingly, v0.50 installs and runs on W2k. I've found no installer for a more recent engine. Delete the Tmp folder between the languages. Aspell6 is a fork by Lyx with a different file format; its languages install on W2k, but without an engine I can't tell whether they work. Imagemagick converts, processes and optionally displays images, it's used by Lyx and others. v6.3.9 Q16 "Windows static" installs on W2k and seems to run, a "Windows dll" exists too. v6.6.4.9 is said not to work on W2k. Lyx v1.6.5-1 Bundle installs Imagemagick v6.5.4-3 Q16 (or possibly a subset) and seems happy with it.

The labor reform that triggered violent protests (could already be called "riot") was voted. Riots aren't so common. There are very few new refugees in France and they aren't violent, but the relations between the religions has degraded brutally after several attacks - it's not so visible in daily life, but quite on forums and newspaper blogs, where the 10 to 30% of openly racist people make 90% of all comments. This is definitely a worry for the moderation since the French law has a bizarre formulation to forbid racist speech but tries to keep the freedom of speech, and this formulation needs non-obvious interpretation. Presently, newspapers use to (under-) pay cheap French-speaking labour forces abroad (typically in Africa) to do some moderation. Though, racist comments are still there, while telling "I got signs announcing this and that hence it was an attack rather than an accident" lets my message disappear and ban my IP. Anyway, I still have to hide my IP and my name to pass through, and up to now Tor does it properly.

I had my own name triggering an automatic (and pirate) ban before. This is NOT a matter of moderators. Call it a "governmental agency" if you wish to distinguish from the government. Answering "error 503" automatically is not what the online newspaper would do to ban a user. This was done by some malware, because at that time, the same message would apss through if I used a different name. Other readers from Germany could still post comments, just as I can with Tor which has IP addresses outside France. Sorry, this case is so blatant, I experimented as a scientist does and observed it so clearly, that it is just plain censorship and nothing else. Discuss Tor without politics? Without politics, we wouldn't need Tor. ---------- More recently, I checked that reader comments were also nearly banned at other French online newspapers where I don't usually put comments, whose web servers are distinct from Lemonde, and where the amount of comments had dropped to about zero. At that time, the Belgian online newspaper had still their reader comments working normally - or more exactly, full of the usual racist messages that would be a legal basis for moderation in France. In the recent days, the situation in France has improved somewhat, especially as I could make the censorship public.

Here are some C compilers, possibly C++, chosen to be simple to install and use, not necessarily with an IDE nor producing executables with a GUI. In other words, you type a Thing.c with a text editor, call Compile Thing.c in Command.com to get a Thing.exe which you run in Command.com to get a few precious numbers on your scren. Borland Cpp v5.5 is free and runs on W2k. The binary is about 5x faster on simple integer operations than the competitors I tried, but Microsoft is 3x faster than Borland - bad sign for the competitors and for the ability of a Core 2 to run unoptimized x86 code. Digital Mars C/C++ v8.57c runs on W2k. Tdm-Gcc is not exactly GnuCC. v5.1.0.3 runs on W2k. I believe it offers an IDE, and for sure, its gcc.exe can work without. OpenWatcom C/C++ v1.90 runs on W2k. This one has an IDE. Orange C/C++ v5.73.1 fails to run on W2k after an alleged successful installation.

Censorship gets only worse in France. The French interior minister had written to his German counterpart (...as I live in Germany?) to find common measures about Internet security since national ones would be inefficient, and apparently the German minister didn't follow. Since then, about all comments on blog.lemonde.fr are censored away. Not only the many racist ones - which France's own version of free speech doesn't make legal -, not only my comments - which were targeted specifically for completely other reasons -, just nearly all. The number of comments has dropped by 100, as confirmed by observation and by the automatic comment counters. What remains is one or two "I like your blog" and nothing else. Tor shouldn't be understood as the eternal answer to censorship, in France nor elsewhere. If the team behind it can't resist the pressure by the governments, Tor will become unsafe, probably even the older versions of the part installed on users' computers.

More worries: Paypal has switched to a stream cipher for its login. It's not exactly the GCM mode but one said to give a better guarantee that the initial vector is unique, oh good. Knowing that A stream cipher is inherently more dangerous than the block mode or chained block; It's obviously as slow as a block mode - whatever nonsense Wiki repeats about AES, it needs one block encryption per data block too, I just wonder WHY so many programmers switch to stream ciphers, including now Paypal which had made sensible choices up to now and had been virtually immune of the attacks that broke Wanadoo, Yahoo and the others. Or do they all give in to the orders of gov' agencies that want to spy everything by collapsing the individual and collective resilience to any attack, including a foreign one?

FreeCad v0.15 and LibreCad 2.0.7 don't. Their installer accepts to run on plain standard W2ksp4r1, but the application stops because Kernel32.dll doesn't suffice. Though, their source is available, so recompiling for W2k may be possible; I tried only the binary distribution for Windows. Candidates for KernelEx? QCad is a partly free 2D Cad simpler to learn than most competitors. Available as open source, so recompiling for W2k may be possible; I tried only the free binary distribution for Windows, but non-free Win binary distributions exist too. v1.5.1 runs on W2k (and on 64b Seven too) while v3.9.3 fails to run after install: "not a valid Win32 app". TurboCad v9.2 runs on W2k. It was cheap, now the editor is closed. Decently capable and simple 2D Cad with "3D extensions", which uses to mean that 3D is clumsy. XnView v2.32 runs on W2k. I tried only the simpler version, as the more complete one targets real photo enthusiasts and has huge menus and tunings. RunAsDate v1.05 and v1.40 run on W2k. RegFromApp v1.15 runs on W2k (but not on W95b). Vilma Registry Explorer v1.6.0 runs on W2k (but not on W95b). It's a better Regedit, faster, easier. Alternate stream v1.51 runs on W2k and many more - as usual with Nirsoft. CCleaner v3.28 runs on W2k while v4.12 doesn't. That's consistent with the editor's claims of v3 for W2k and with the installer's messages - kudos. GnuPG's latest v2.3.2 and v2.1.14 run on W2k - these have binary distributions for Windows. The source code is available, so compiling other versions for W2k may be possible. GnuMeric is a free spreadsheet compatible with Excel with better menus for diagrams. v1.12.17 runs on W2k and is the latest binary distributed for Windows (201406) by the authors, alas. It's also the best version: keeps old bugs but adds nice features. Compiling more recent ones must be possible. BaKoMa is a complete and not very expensive distribution of LaTeX for Win. v9.05 runs on W2k and is said to be the latest. Apparently it needs a disk volume where Win is not. MuPDF displays Pdf documents damned fast but does nothing else - not even text search nor jump to page. v1.7a runs on W2k while v0.7 fails (in that order, I confirm). SumatraPDF adds menus to the MuPDF viewer. v3.0 fails on W2k after installing, (latest v2) v2.5.2 refuses to install, (latest v1) v1.9 refuses too, and (latest v0) v0.9.4 runs but is slower than Foxit on old documents scanned as pictures.

Hello dear friends! Internet is ever more heavily censored in France. Of course it began with child pornography, drugs, explosives and so on, and has drifted as expected. The country has arrived to a point where free speech isn't possible any more. Le Monde, a traditional big serious newspaper, has an online edition and a blog section open to reader comments lemonde.fr and blog.lemonde.fr In April 2013, the server began pour "crash" when I posted comments unpleasant for the French state. Which wasn't like "make revolution" but rather "The study by Prof Seralini about Monsanto's GMO is statistically not so weak". I experimented around it, and found very clearly that my real name, combined with some keywords, triggered the censorship. I could go on using some simple measures. Since 26 July 2016, I couldn't drop any single message more, even when using an alias. My last message was approximately "I got messages suggesting an attack before it happened in Reutlingen, so the authors were a secret service, probably a French one, not Daesh". Supposing that the server censored my IP, I installed Tor Browser https://www.torproject.org/projects/torbrowser.html.en#downloads and, provided I use an alias, can again put messages, at least on the blogs that aren't moderated before any publication. Though, my messages where I give the link to Tor Browser are suppressed - in one case, only the paragraph linking to it. To my best knowledge, using Tor is legal in France. Spooks tried in December 2015 to let forbid it but the government didn't. Having a long history of conflict with the French spooks, I'm not surprised that the country evolved that way, but you may be interested in reading that it happens in decently reputated countries too. And I encourage everyone to download and store while still possible these tools that enable free speech and protect our privacy. Presently (it got public at end July) Tor has problems with a sex scandal, so I feel prudent to download the present version (6.0.3) before the team is coerced into botching the software, or even, prefer a slightly older one. Gratitude to all these people helping us protect our freedom.

What version of Tor Browser runs on plain vanilla W2k? After trying, v6.0.3 and v6.0.2 don't. Their installer accepts to run on plain standard W2ksp4r1, but the application stops because Kernel32.dll doesn't suffice. Candidates for KernelEx?

Thanks GrofLuigi! I have tried RMClock v2.35 meanwhile. It starts cleanly on W2k and displays much meaningful nformation about my Wolfdale. I was pretty sure that my Cpu's clock stuck at 3.33GHz previously. Since the first time I ran RMClock, and without asking anything special nor seeing anything that would start it a bootup, my Cpu now throttles nicely to 2GHz when less loaded, both on W2k and on Xp - before I ran RMClock on Xp. So either I misread the clock previously, or RMClock made something.

Hello nice people! My Cpu has the Speedstep feature but I didn't find where to enable it. Observations so far: Intels says my E8600 ("Wolfdale" core 2 duo) features it. http://ark.intel.com/products/35605/Intel-Core2-Duo-Processor-E8600-6M-Cache-3_33-GHz-1333-MHz-FSB Cpu-Z observes 3.3GHz all the time. The mobo is a Ga-ep45-ud3r with P45 northbridge, bios upgrade F11. I didn't find any setting nor mention in the Bios. Photographs available. The OS is Xpsp3 (and W2ksp4r1 but that's less important) with vey few updates. I didn't find any mention in the Control Panel. The cpu driver is Intelppm.sys v5.1.2600.5512 hence brought by Xpsp3. I've downloaded RMClock v2.35 but still not tried it because it doesn't mention the Wolfdale processors. Hints, proposals, suggestions, explanations...?

I take good notice. I had a single mobo with Via Pata host which did corrupt the data. At that time and also with later Via hosts, including at the beginning of Sata, many more users noticed data corruption. Since this is THE unacceptable failure, I decided never to have a Via host again. But I did have Via Usb 2.0 / Pci chips that worked nicely. As good as Nec.

I did on previous computers, especially a Pata Win95 laptop, and experimented with W98-Me-2k-Xp. It does bring a lot. It is not obvious, and you can waste a lot of time. Unless you really want to stick to a Pata computer, just buy a Sata mobo and put an Ssd on it. Or put a Sata host in a slot (SiI3124 is by far the best for Pci and also the cheapest; take any board brand) and connect an Ssd. More GB, cheaper, works without the worries. So let's imagine you have no choice and really need a CF. Only one brand and model used to work properly, that was Transcend 300x and 266x in 1GB, 2GB, 4GB and 8GB sizes, not 16GB. These ones were SLC: MLC is MUCH slower and survives fewer writes. Though, Transcend now counterfeits itself and sells bad MLC 300x 8GB, so avoid them. msfn.org/board/topic/171331-fake-transcend-compact-flash-card/ I don't know if the 4GB are still good. You may have better chances with the 266x, which are about as fast as the 300x anyway. Besides SLC, which would be available from Adata and others too, you need buffers working at 5V, and Transcend does this, others not often. CF standards accept 3.3V but Pata does NOT, and is talken in 5V, most CF cards slow down horribly to minimum Uata or even to Pio, and then you have lost, or corrupts data, and you have lost too. On the other hand, the Pata cable and the adapter are unimportant, and you can have two cards on one cable and adapter (worked in my laptop). To install Xp (and I strongly believe 2k too) the card must show the fixed bit. Only some models do it, noticeably the Transcend 266x 300x, Adata and some more. But W95 doesn't need it. Install Windows normally with the Cd, unless you make special experiments by copy-paste. The Windows install Cd checks the integrity of the installation, and this is the best test of your hardware. If the installation fails, don't waste much time with settings, driver choice and so on: it's a very probable sign that your hardware combination does not work, and if you install by copy-paste you WILL have trouble, only later, when you precious documents are on the machine. Failure to install uses to result from the 5V buffers. If you can have several CF in your machine, you could maybe perhaps have a big SLC for files written seldom, like the applications, but NOT your documents, the OS, the paging file, the browser caches. But on an SLC, put the paging file and stay relaxed. I already had raid/0 della muerte (2* Transcend 300x totalling 90MB/s) on a Tualatin+i815ep+Sil0680 (the best Pata on Pci chip) and it brought nothing faster than one CF. Importantisimo: align the volume on page boundaries. Vital for speed, important for the write amplification. With Ntfs you just align the volume on MB, which Windows 95-Xp don't do but GPartEd does. With Fat32 (if you have to) it's more difficult msfn.org/board/topic/151798-does-fat32-align-its-clusters/#comment-968582 If your hardware or Bios limits the disk size, or accesses only through CHS, or can boot only from sector 63, or needs a hibernation file (like my laptop), expect serious trouble.

Sounds reasonable. I had just hoped to find sometthing as easy to install and use as Javascript on Firefox, which was a good surprise. I'll read the list, try to pick the proper compiler and install it. Thanks!

At the Pata era, Via was unable to produce a working host chip. All their chips corrupted the data. It was not a matter of tuning, driver, whatever - just flawed chip design. The best explanation is that their Sata chips too are just as flawed, and Sata1 likely isn't much better than Sata2 setiing. So just keep away from any disk host by Via: these people don't know how to design them. But have Via Usb 2.0 chips, these work. Put your disks on the Promise chip only. And use Sata mode on it, at full throttle, because with Pata you lose the simultaneous and out-of-order multiple requests, which do make a difference. More so with mechanical disks, but with Ssd too.

Thank you once again, Jaclaz! Downloaded, installed and briefly tried. It does run on W2k, and claims to works with W95, nice. It does already more than what I sought but it does it well and is quick to learn. Very seducing. Maybe I use it for htm instead of a text editor. Other suggestions still welcome!

Thanks Tripredacus! That's clearer. And is there any known weakness in the GCM mode of AES provided by Firefox? I've seen it's a pseudorandom sequence mode (sequence xor plaintext = cyphertext), which is **** difficult to program without introducing weaknesses. For a month, Yahoo uses a new login page, where I first type my identifier, and then the page refreshes and I type my password. So if the pseudorandom sequence is somehow linked between both uses by the same page, for instance if it's identical, the user loses. An attacker knowing the identifier and hearing the connection deduces the pseudorandom sequence then the password. I don't see why programmers for encrypted pages go for pseudorandom sequences. Normal modes like codebook or chained codebook are as easy and not so horribly dangerous.

Hello everyone and everybody! When I visit an https website where I provide data that must be encrypted (identifiers for instance), who provides the encryption, among my browser and the website? I attach an example with Firefox used to log at Yahoo. I vaguely imagine that the browser (relying possibly on the OS) provides cryptographic services, the webpage tells "please encrypt the user's identifiers" and "I understands these codes and protocols", then the browser looks for a set of codes and protocols spoken on both sides, then they define and exchange some keys and transfer the data. Is that it? Thank you!

Hi the experts! I've programmed in C and C++ (and older ones) few decades ago, and now I consider writing just one small and dirty code, like 50 lines of recursive integer computations. What I need: Accurate integers with bitlogic, on 64 bits better than 63, but running on 32b Xp and preferibly W2k, and at a good speed, similar to compiled. Apparently, Javascript limits to 50+ bits and Java to 63 bits or even less on 2k-Xp. I don't need an object-oriented language, nor an IDE to develop the code. Text editor <-> Line compiler <-> Console would be OK. But free please. The code is for me, by me only, no documentation planned, and it will run on the developing machine. No GUI for the application, single task, no Avx: things like printf(). I want to spend very little time on installing the interpreter or compiler and learn it and the language. You known, the source libraries, the parameters for W2k target, the runtime links... Javascript was a good surprise in that aspect. So: which one do you recommend? Thanks!

Hello dear friends! Up to now, I've made my htm files with a Wysiwyg editor (Namo), and presently a general text editor (Metapad) for better control of the code. I plan to stay at a text editor but would gladly accept some help from it, like highlight the htm keywords, maybe insert tag pairs and special characters. But not much more: for instance, no automatic saving of images. It must run on Win 2k and Xp, if possible 95b. Which editor do you recommend? Thanks you!

In 3 messages from you I didn't get the answer. I asked "How do the programmers make use of the Sse and Avx" and "what does the programmer need to do, and how autonomous is the compiler?", and got vague and out-of-topic recommendations at the level of a general culture course. Reading the same material in a Press paper I'd regret that the journalist only reproduced some sentences he grasped during a Powerpoint presentation. Meanwhile I got the sought information through varied websites, for instance http://svmoore.pbworks.com/w/file/fetch/70583970/VectorOps.pdf http://sci.tuomastonteri.fi/programming/sse http://felix.abecassis.me/2011/09/cpp-getting-started-with-sse/ In case this is helpful to someone else: The programmer can obviously include assembler in his source code. Example here http://neilkemp.us/src/sse_tutorial/sse_tutorial.html which has clear drawbacks, hence the desire for alternatives. Most compilers define extra data types corresponding to the Sse or Avx data packets, and functions that operate on these data types, overloading the usual operators for some functions. It needs a few more conditions ilke data alignment and special keywords. The most convenient is when the compiler recognizes that a portion of source code can run on the Sse or Avx, because it keeps the source easier to read, and an existing source can be recompiled for more recent processors -with luck, present code could even be recompiled in the future when processors offer new possibilities. http://svmoore.pbworks.com/w/file/fetch/70583970/VectorOps.pdf on page 8 calls "Automatic vectorization" this autonomous work by the compiler, which isn't widespread, works more or less and can bring worries. Intel's Icc compiler does that and was possibly the first one. Visual Studio 2012 and later try to do it, Gcc too. Over portions of the source that obviously can run on the Sse or Avx, their success rate varies a lot. Vs and Gcc create only the binary for the intended target machine without adding tests for the adequateness of the actual processor. This binary crashes if run on a processor lacking the instructions. Intel's Icc does some tests, but obviously it would have been the programmer's role and is often neglected. Not specific to compilation: Having used Avx256 instructions can make Sse code, or scalar code that runs on the Sse as on the Core, much slower. Only the registers provide the data throughput to feed present compute hardware. The L1 cache suffices in some cases, other caches and the main Ram never with existing processors. The Sse and Avx make that worse. Compute loops must be deeply reorganized according to the caches' possibilities; compilers don't do that alone, so it remains the programmer's duty.