VistaLover Posted February 21 Posted February 21 14 hours ago, modnar said: Good point about Googl-isation and it's a nasty thing, in my opinion largely caused by Adobe's incompetence (?) to keep Flash together Like you, I do agree that Adobe Flash Player, despite being closed-source, was a good technology for its time and served well XP/Vista users running the OS under era-correct H/W; it came embedded with patented decoders and worked quite well with most GPUs of the era... But it wasn't deprecated due to "Adobe's incompetence", nor was this related to just Google pushing their ways... FWIW, a localised version of Flash is still being used in mainland China, maintained by a Chinese Adobe affiliate... As Microsoft moved away from XP (and Vista) and introduced new media playback technologies inside the Windows OS itself (e.g. Windows Media Foundation, WMF, which matured in Win7), and with newer WinOSes requiring more "advanced" H/W (GPUs/CPUs/RAM) to run properly (but through which, higher-spec'ed, H/W native playback becomes "attractive"), browser vendors saw it proper to move away from NPAPI/PPAPI Flash into HTML5 native playback, inside the browser; Adobe Flash had been notorious for many serious vulnerabilities during its lifetime, so for many computer users the deprecation of Flash was a blessing in disguise... Google was not the only browser vendor in favour of the migration, in fact Apple had a bigger role in this ... I can understand, though, that the deprecation of Flash and the general move to HTML5 media playback has left XP users on older/under-resourced H/W at an impaired/handicapped state.... But when I talked about the "Googl-isation" of the web, I did not have the deprecation of Flash (and NPAPI in general) in mind; I meant the current trend of Google web devs constantly "devising" new "dialects" of the Javascript programming language and them, via their undisputed dominance on everything-web, leveraging these new JS iterations into the web standards; these, even when still at "draft" status, are pushed to the newer Google Chrome releases and, as a result, become also adopted by the web-frameworks that are part of most contemporary web sites; sites that, unlike the good times of the past, aren't just comprised of pure HTML pages, but are instead monstrous blobs of JS+CSS code that has to be downloaded and rendered locally by the browser engine, putting a heavy tax on your H/W... Thus, Google make sure that "the web" will break on other "legacy" web engines and also make sure that "the web" will underperform on your older H/W - IOW, "they" try their best to make your old "setup" practically no longer capable of any more use ... People who follow me here know that I'm not a coder/web developer, so it's possible some of my wording above was not well chosen; but I'm sure you all get the gist of it ... 4
VistaLover Posted February 21 Posted February 21 19 hours ago, reboot12 said: The change OS value Windows NT 10.0 to Windows NT 5.2 in UA probably helped: Mozilla/5.0 (Windows NT 5.2; rv:120.0) Gecko/20100101 Firefox/120.0 Hi again ; I can't see myself how that could've helped , because Fx-120.0 is incapable of running under NT 5.2 (Windows Server 2003 x86 ?); with the above UA, you're just "sticking out like a sore thumb" ; still, stranger things have happened in the world of IT... Also, if elektroda are adamant on supporting only the latest Mozilla Firefox version, pretty soon you'll have to spoof Fx-123.0 ... 20 hours ago, reboot12 said: Thanks for showing interest ... And you're most welcome ... 4
reboot12 Posted February 21 Posted February 21 (edited) @VistaLover Unfortunately, the change of Windows NT 10.0 to Windows NT 5.2 did not help and Secure... appeared again Changing the value to Chrome also does not help. When I have a Chrome setting and I delete cookies, I can enter the forum main page but when I try to log in, there is a Secure Connection Failed error If I remove the value general.useragent.override.elektroda.pl and cookies then always Secure Connection Failed appear but after restart browser possible enter to forum and log in and this only works for a moment about 3 clicks on the forum and Secure Connection... For now, the best option is UA: general.useragent.override.elektroda.pl Mozilla/5.0 (Windows NT 5.2; rv:120.0) Gecko/20100101 Firefox/120.0 There is no problem around one day or several hours. Edited February 21 by reboot12 1
mina7601 Posted February 21 Posted February 21 18 hours ago, VistaLover said: NT 5.2 (Windows Server 2003 x86 ?) Yes, it's detected as Windows Server 2003 x86.
VistaLover Posted February 21 Posted February 21 (edited) On 2/15/2024 at 2:38 PM, Egorkaru said: https://web.telegram.org/a still does not work in Serpent 52 and New Moon 28 browsers Greetings ; what do you mean by "still" ? Was it working in UXP-based browsers and then stopped? How long ago was that? Have you already reported that breakage in the past and you now feel disgruntled the issue remains unresolved? Please be aware that UXP, as a platform, is not very good at supporting these so-called web (in-browser) editions of popular chat/messaging applications, which are usually optimised for recent Chromium-based browsers (or just for the browsers on your mobile device ) ... Such web-compat issues are usually for "upstream" (MCP) to address, but if one asks "there", the standard answer from "them" is: "Use the dedicated Windows application"; this is, of course, no consolation for XP/Vista users, because most of these apps now require at minimum Win7 (soon to be Win10 ?) ... Searching in the PMForums, someone there suggested using a "legacy" URI of the form: https://web.telegram.org/?legacy=1#/login , does it still work on you for log-in purposes? On 2/16/2024 at 10:20 AM, Egorkaru said: a redirect occurs to a stub page about an unsupported browser: https://web.telegram.org/a/unsupported.html ... In my "dirty" St52 profile, all I get is a white/empty page inside a thin-lined black frame , much like what @mina7601 has already posted... On 2/16/2024 at 10:20 AM, Egorkaru said: from two errors in the console: Quote Content Security Policy: Couldn’t parse invalid host 'wasm-unsafe-eval' Load denied by X-Frame-Options: https://web.telegram.org/a/unsupported.html does not permit framing. It would appear the most important clue inside the Web Console simply eluded you ; the page deploys a script which performs browser-feature checks: https://web.telegram.org/a/compatTest.js UXP then fails those tests , because it doesn't, yet, support "Intl.DisplayNames", a JS feature which was first implemented in Fx-86/Ch-81: It just so happened that "upstream" (MCP) have opened a UXP issue for that just 6 days ago, but so far it hasn't received any activity ... A polyfill for that missing function already exists ; my JS-coding skills are non-existent , but I tried to make a userscript out of it (and inject it through Violentmonkey), problem being the script gets quite large when you add all the needed "locale.data" ... Another issue is that, during my tests, I discovered that for the userscript to work on "telegram.org", one has to globally disable CSP protection in the browser, which. in itself, is a security risk ... FWIW, with my basic "polyfill-as-userscript" and CSP (temporarily) disabled (security.csp.enable;false), the offending web page finally loads: I don't own a mobile device (yet anyway , by my government has other "plans" for its subordinates ), nor do I have an account with Telegram, so my experiments ended just there ... Kind regards. Edited February 21 by VistaLover 4
mina7601 Posted February 22 Posted February 22 2 hours ago, VistaLover said: does it still work on you for log-in purposes? It does, but the problem is, since it's legacy, it won't load messages from newer Telegram. 1
Jody Thornton Posted February 22 Posted February 22 On 2/19/2024 at 6:53 PM, VistaLover said: Dear Jody, and I mean this in a genuine way , you of all people shouldn't come back with the very same queries, because it was I who answered your similar queries in the not so distant past... What it all comes down to is: 1. do you have a grasp of what open source code is? 2. do you have a grasp of what a forked open-source code repo is? 3. do you have a grasp of what a platform and an application built on it is? If not, any answer you'll get from me won't make much sense to you, and I'm sorry to say that there's no simplistic yes/no answer to your re-iterated queries above... I will re-read it, but you must understand that I also get things change, and some people (me for example) are just too busy to read through pages of threads. Sometimes we also just forget. So I don't feel bad about being "ignorant" about it. I've learned to ask as I please, and not feel badly about it. There's nothing wrong with re-visiting a question. It's not meant to inconvenience people in the least. That I promise. And I do thank you for the explanation. 1
Jody Thornton Posted February 22 Posted February 22 Vistalover Wrote: This is a fork of the official UXP platform repo (see above); the tracking branch of that repo follows more closely the master branch of the official UXP platform (see above); the forked UXP repo, by now, is different to the original one in various ways, one of which is in restoring WinXP+Vista support (which also entails several lib differences, like in ffvpx), another one is keeping Mozilla features MCP have dumped long ago (e.g. Web Extensions, Tab Containers, half-baked e10s code, EME, e.a); also, the NSS lib in "our" browsers is somewhat different to the one MCP maintain; that is why profiles between the official apps and roytam1 apps aren't 100% interchangeable... So see, I'm glad I asked again because when I started following this years ago, I really thought that New Moon 27 was just Pale Moon 27 modified to run on XP and Vista, and that every update since then was the same comparative. Likewise for Pale Moon 28 vs New Moon 28. I DID realize that never releases kept the 28x versioning, though in fairness to myself, I DID go and look in the first posts of this thread where we're told that was explained, and I didn't see that explanation (I I only add that point to show I did research again before asking)
Jody Thornton Posted February 22 Posted February 22 On 2/20/2024 at 5:05 AM, mina7601 said: Done. Agreed! As for feeling a sense of "responsibility" for keeping up with changes, I don't 100% agree with that. I do feel that there is a valid option of just wanting being a "consumer" of things. I don't need to know how my toaster works, and if I feel a need to ask someone once a year because the interest strikes me, I'll have at it. That may seem argumentative to some, but it's not intended to be at all - simply an observation and point of views, and a perfect valid one at that. Changelogs contain LOTS of content and many may be too busy to read through that. Likewise @VistaLover, you also have no responsibility to answer my questions, and while I really appreciate that you did, I won't be made to feel like I I somehow put you out because I asked. That's not meant to be confrontational, so I hope it's heard in the spirit that it's meant. 1
shelby Posted February 22 Posted February 22 Big tech builds the AI so the old machines go for scrap. That is why the internet change the form to IoT. Welcome to New World Order... 2
modnar Posted February 22 Posted February 22 2 hours ago, shelby said: Big tech builds the AI so the old machines go for scrap. That is why the internet change the form to IoT. Welcome to New World Order... We surely need to work against that; any way we can. 1
we3fan Posted February 22 Posted February 22 (edited) On 2/21/2024 at 11:38 PM, VistaLover said: https://web.telegram.org/?legacy=1#/login , does it still work on you for log-in purposes? I tested this few days ago on 360Chrome v13.5 : https://web.telegram.org/?legacy=1 - login with phone number only, doesn't log me in when I enter my code hmm (tried 2 times) https://web.telegram.org/z/ - redirects to web.telegram.org/a https://web.telegram.org/k/ - doesn't show QR code hmm; later showed QR code but it took too long (1-2 min maybe) https://web.telegram.org/a/ - login with QR code or phone number : 360Chrome v13.5 - I can only send messages, but I can't call, call tries but can't connect hmm GDIChromium 103.0.5060.122 on Win 7 - call worked but it had loud audio glitches hmm Apologies if Chrome and Win 7 are OT here. EDIT: Just wanted to share my web.telegram.org experience, I happened to test it on Chrome. Edited February 23 by we3fan 1
mina7601 Posted February 22 Posted February 22 (edited) 6 hours ago, we3fan said: I tested this few days ago on 360Chrome v13.5 : https://web.telegram.org/z/ - redirects to web.telegram.org/a It does that even on the newest Chrome, version 122 at the moment. So, nothing to blame on the browser, but probably on the incorrect creation of the website. 6 hours ago, we3fan said: https://web.telegram.org/?legacy=1 - login with phone number only, doesn't log me in when I enter my code hmm (tried 2 times) Logs in with code just fine on 360Chrome v13.5. Are you sure there's not some extension interfering with the login process? Check with a new profile to make sure! 6 hours ago, we3fan said: Apologies if Chrome and Win 7 are OT here. No, no, no apologies needed. Edited February 22 by mina7601 2
VistaLover Posted February 23 Posted February 23 (edited) @Jody Thornton : No worries, we're good ... You're right, nobody "twists my arm" to make long-winded posts here, but in your case I did provide replies to your queries because I thought you deserved the replies - also, this is material that, no doubt, will be of interest to other members here (and see, roytam1 pinned it to the start of this thread) ... 22 hours ago, Jody Thornton said: I really thought that New Moon 27 was just Pale Moon 27 modified to run on XP and Vista, A correction on a technicality there: "official" Pale Moon 27 had native support for Vista (the 64-bit variant had some glitches with media playback - only recalling from memory, because my Vista OS is 32-bit), so NM27 by roytam1 only had to re-instate WinXP support ... 22 hours ago, Jody Thornton said: Sometimes we also just forget ... Tell me about it ... Old(er) age is a b*tch ... 22 hours ago, Jody Thornton said: I don't need to know how my toaster works ... Neither do I, but I'm the type of person that first reads the toaster's manual (if it comes with one ), and only then do I plug it to the mains; do you get my drift? 22 hours ago, Jody Thornton said: I DID realise that newer releases kept the 28.x versioning, though, in fairness to myself, I DID go and look in the first posts of this thread, where we're told that was explained, and I didn't see that explanation ... well, sorry to say it, but you didn't look well : On 1/20/2024 at 1:14 AM, roytam1 said: Q: Why you still keeping on version 28? will you have a version 29 build? Linked Answer: https://msfn.org/board/topic/182647-my-browser-builds-part-3/?do=findComment&comment=1199724 If I'm being asked (once) again , the decision back then was made based on uncertainty on how several, then popular, Firefox extensions would behave under an application with version 29.0+; it was a time when "upstream" (spearheaded by M.A.T.) wanted to completely remove ALL support for Firefox-specific extensions and move solely to PM targeting ones (in fact, they did do that for a while, didn't they?) ... The transition from v28.x to v29.x was crucial for Mozilla Firefox, because Fx-29.0 came with radical internal and external changes, collectively known as Australis (which, contrary to public belief, wasn't just a change in GUI); many Firefox extensions of that era, to keep backwards compatibility, had different sets of internal code to cater to Fx <=28 and Fx >=29 at the same time, so using such an extension on NM29 (with a Fx-28 type of GUI/engine) was just asking for trouble... Edited February 23 by VistaLover 1
VistaLover Posted February 23 Posted February 23 9 hours ago, we3fan said: Apologies if Chrome and Win 7 are OT here ... Well, Win7 is certainly "On Topic" here, because the roytam1 forks (browsers and other apps) do work under that OS; after all, despite its users not wanting to admit it, Win7 is by now an "Older NT-Family OS" ... However, and this is just me , "extensive" references to the Chromium forks (compatible with said "older NT-family OSes") should be better posted in the dedicated threads (we have enough already...) ; I don't exclude myself here , I've often fallen foul of this before ... 2
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now