00:13:32 tomman: imagine if the web browser had the capability of render content itself 00:14:12 s@render@&ing@ 00:14:13 Network calls to download a static post: 2 (one for the post, another one maybe for the CSS) 00:14:32 network calls to download this abomination: at least 10, including to offsite resources 00:15:35 plus all the extra burned CPU cycles, the inconvenience of breaking your site on browsers not approved by The Junta, and even the initial weird rendering artifacts on Compliant™ browsers 00:15:42 back when more sites I browsed at least had content properly in the html, disabling css and js in a mobile browser was like installing windows 98 on a DDR2 machine with an SSD 00:31:02 Windows 98 on a SSD is a nerd dream come true 00:31:25 those that argue that the charm of retrorigs are slow-as-molasses dying rust spinners can go choke on a Conner Cabo 00:31:40 this is why every single of my retrorigs has solid storage :) 00:33:39 there's probably a rabbit hole to follow if one is interested in that, trying to see how much can be modified to also improve performance overall, I suppose some have at least done some stuff like that, wasn't there a third-party proprietary driver to allow more RAM? or maybe I'm confusing with something else 00:34:35 (installation IIRC is possible, although easier if you can replace with a, say, 512 MiB module, and what was the ceiling with the stock OS, 2 GiB?) 00:36:55 9x/Me is limited to 512MB on a stock, unmodified system - anything over than that will trigger icky BSoDs on boot unless you edit SYSTEM.INI to cap max physical address below 512MB 00:37:26 you can install it, it will just go burst in flames after the first reboot 00:38:09 there are 3rd-party patches to lift out this limit, the most famous one is the RLoew's PATCHMEM (which used to be payware, until the dude died and his family released their patches for free) 00:38:36 and EVEN THEN you may find out machines with weird errors still requiring SYSTEM.INI edits (like my dualboot T40) 00:39:18 (had to declare the final 4MB a no-man zone or things like WMI, Clippy, and some DOS boxes would work erratically or not at all) 15:48:01 WOWOW, I finally got a stupid Turnstile checkbox to check on Seamonkey! 15:48:27 I had to endure the usual ~270 seconds timeout on my Ryzen, but this time the goddamned checkbox did actually rendered AND checked AND let me in 15:48:56 so... progress? 15:49:37 at least this time the checkbox didn't timed out, but I still have to poke at the box with the Inspector, fall into the endless script loop, and wait 16:03:41 tomman: wow 16:03:55 didn't know 270s is a magic number there 16:04:05 should try that with the very old icecat mobile thing. 16:04:25 (unrelated to iceape) 16:04:40 Harzilein: it's hardware depending 16:04:52 on my Ryzen 7 5700U is ~270 seconds 16:05:00 were there ever any attempts to make "uxp mobile"? 16:05:01 on a old Core 2 Duo T7200 is... 20 minutes 16:05:20 should benchmark this Sandy Bridge one of these days, but I would expect no less than 10 minutes 16:05:29 and by then the captcha has already expired 16:06:13 is there something human readable that indicates expiry? or is it just a page refresh and vague notion of "notify us if this fails" or w/e? 16:06:35 basically 16:06:45 if you're real lucky, you MAY get a error message on the actual captcha box 16:06:54 in red letters 16:07:01 turnstyle slot machine 16:07:05 but most of the times nothing renders and the page just reloads after the browser un-hangs 16:07:12 I did got somehow lucky today 16:07:20 TurnSickle™ 16:07:46 yeah, you got drawn in the "let this loser access his warez download" sweepstakes ;) 16:10:12 i mean, is there any notion of them not considering their clients' clients as filthy unwashed masses in the first place? not potential, everyone, just some get some mercy. 16:11:01 science.org is fiddly 16:11:08 sometimes you get Clownflare'd 16:11:11 sometimes it loads fine 16:11:33 (Derek Lowe told once that there were some exceptions setup, but the whole stuff is finicky at best)