|
This reads like someone arguing against FORTRAN in favor of assembly in the 60s.
|
#
?
Sep 12, 2014 19:28
|
|
|
|
| # ? Jun 7, 2024 07:29 |
|
|
Am I reading it wrong or does he think most CSS is used for typography? Suspicious Dish posted:This reads like someone arguing against FORTRAN in favor of assembly in the 60s. I just assume that every time he looks around his bathroom at home he wonders why there's a toilet, sink and bathtub when the tub is perfectly capable of doing the work of all three. E: maybe not the best analogy, but I definitely had the same thought while I was reading it before I saw the date. Munkeymon fucked around with this message at 21:15 on Sep 12, 2014 |
|
#
?
Sep 12, 2014 20:17
|
|
|
eithedog posted:It struck me as hilarious that I've seen a CSS rule spanning over 500 lines. Then I continued on drinking. streetlamp posted:i need to see this piece of art I dunno if this is what eithedog was referring to, however I do recall him enjoying it when he saw this: Westie posted:Also, a site went live today. And uh, things weren't going that easily so I decided to have a look at the selectors to see what was going on. And yes, that was the only instruction - to set the background colour to white.
|
|
#
?
Sep 12, 2014 21:41
|
|
|
Munkeymon posted:Am I reading it wrong or does he think most CSS is used for typography? He thinks that it's 20 years ago. Another thing about it, is that it's not even valid HTML. Where are his closing tags? Not one </p>, no </html>  streetlamp posted:i need to see this piece of art Generated in LESS - somebody forgot what a given macro does and how does it expand, and didn't notice until we looked at the loading times of the page and why this CSS file weights... some (did any of you see CSS files that weighted ~1MB?). It didn't help much that the containers had all names like "div.communications-module-1-container" or that the CSS folks used it as "body.main-body-container > div.main-content-container > div.left-sidebar-container > div.left-sidebar > div.left-sidebar-first-menu-container" etc. Unfortunately it ceased to be, but don't worry, next time (and there definitely be a next time) I'll save it to wonder what causes people to do that. I don't know if it's hate or just not bothering. canis minor fucked around with this message at 21:57 on Sep 12, 2014 |
|
#
?
Sep 12, 2014 21:50
|
|
|
ExcessBLarg! posted:There's also the problem where the world runs on a shitload of legacy software. Can you run a legacy C library in this environment? If so, does it still protect you against all the traditional C-language vulnerabilities for which we've spent decades hardening traditional kernels? If either answer is no, yeah, you're not going to get wide adoption. True enough; I thought it would be interesting to support running legacy code or VMs in this kind of environment, but make them live in ring 3 with traditional memory protection and have a ring 0 VM manager/proxy handle messages on their behalf. Not that I have the time, knowledge, or inclination to write an operating system... I just downloaded the Singularity public release and played with it in a VM and poked at the source a bit.
|
|
#
?
Sep 12, 2014 22:09
|
|
|
eithedog posted:He thinks that it's 20 years ago. Another thing about it, is that it's not even valid HTML. Where are his closing tags? Not one </p>, no </html> That's valid HTML. Closing tags of certain elements are optional, and that has been part of the spec for a long time.
|
|
#
?
Sep 12, 2014 22:29
|
|
|
I have a grammar horror to report. What's the opposite of active? If you answered CSS code:And, no, this wasn't a misspelling of 'deactivate', it's used as the inverse of .active but good thought.
|
|
#
?
Sep 12, 2014 22:37
|
|
|
Suspicious Dish posted:That's valid HTML. Closing tags of certain elements are optional, and that has been part of the spec for a long time. Well there's a lot of valid stuff that isn't necessarily the best way to go about it. Example: bool foo = items.Count > 0 ? items.Count == 1 ? true : false : true ad nauseum. XHTML has it right with closing tags imo. vv And yes Knyteguy fucked around with this message at 23:25 on Sep 12, 2014 |
|
#
?
Sep 12, 2014 22:47
|
|
|
Suspicious Dish posted:That's valid HTML. Closing tags of certain elements are optional, and that has been part of the spec for a long time. The pedant in me will disagree  (to be fair I always treated HTML documents as XML, and, as such it would be an irk to have malformed nodes... - the more you know, I guess)
|
|
#
?
Sep 12, 2014 23:17
|
|
|
eithedog posted:(to be fair I always treated HTML documents as XML) Too bad your browser doesn't, and neither does the WHATWG.
|
|
#
?
Sep 13, 2014 00:38
|
|
|
Ender.uNF posted:True enough; I thought it would be interesting to support running legacy code or VMs in this kind of environment, but make them live in ring 3 with traditional memory protection and have a ring 0 VM manager/proxy handle messages on their behalf. Yeah, it's interesting stuff to think about, it's just too hard to move to towards. I'm actually kind of amused we're doing the opposite now: writing applications against VMs hosted inside strict sandboxes (web browsers) that use a lot of message passing (in the form of REST calls). All of which is built on top of a base of probably-insecure-but-getting-better system and UI libraries that date back 30 years. And they said microkernels were slow.
|
|
#
?
Sep 13, 2014 00:47
|
|
|
eithedog posted:The pedant in me will disagree There was a brief time when I served my website with a Content-Type: application/xhtml+xml HTTP header, which forces browsers to treat the payload as strict, correct XML or else display a parsing error. But even though my website has perfect XHTML markup (even the comments, thanks to draconian input validation), several moderately obscure browsers refused to render it as a web page, and I had to revert it. We live in a fallen world. qntm fucked around with this message at 01:10 on Sep 13, 2014 |
|
#
?
Sep 13, 2014 01:06
|
|
|
Munkeymon posted:I have a grammar horror to report. What's the opposite of active? Similarly, what's the opposite of "plugged in"? Specifically in the context of firmware for a NAS device. If you said "unplugged", as in "Warning: drive 3 was unplugged", then you have a better grasp of English than QNAP's developers. "Warning: drive 3 was plugged out".
|
|
#
?
Sep 13, 2014 01:13
|
|
|
qntm posted:There was a brief time when I served my website with a Content-Type: application/xhtml+xml HTTP header, which forces browsers to treat the payload as strict, correct XML or else display a parsing error. But even though my website has perfect XHTML markup (even the comments, thanks to draconian input validation), several moderately obscure browsers refused to render it as a web page, and I had to revert it. There's always XSLT 
|
|
#
?
Sep 13, 2014 01:27
|
|
|
Suspicious Dish posted:This reads like someone arguing against FORTRAN in favor of assembly in the 60s. I wonder how much assembly was actually being written in the 60s vs fortran/algol (ok, it was a long decade tech-wise, but still).
|
|
#
?
Sep 13, 2014 03:11
|
|
|
Shitloads of both were being written for a long time to come (actually I don't know how much Algol-60 or -68 got written "in anger", though it was used for basically every paper published for years). All the firmware for early digital gauges and transducers that my employer put out in the 80s were done in (Z80, 68HC08, or 6502) assembly. On a somewhat less serious note, NBA Jam was written in assembly (in 1993!), as were most games before about 1990.
|
|
#
?
Sep 13, 2014 03:52
|
|
|
All of the games for the original Nintendo and then Super Nintendo were hand-written assembly, unless somebody wrote a compiler we aren't aware of.
|
|
#
?
Sep 13, 2014 07:40
|
|
|
Suspicious Dish posted:All of the games for the original Nintendo and then Super Nintendo were hand-written assembly, unless somebody wrote a compiler we aren't aware of. And there are some pretty amazing bugs coming out of them, if you follow videogame speed running at all.
|
|
#
?
Sep 13, 2014 10:08
|
|
|
Suspicious Dish posted:All of the games for the original Nintendo and then Super Nintendo were hand-written assembly, unless somebody wrote a compiler we aren't aware of. I remember reading an article about how Sonic Pinball was the first Sonic game to be written in C rather than assembly, about how everyone thought it was a crazy idea at the time, how they don't think it was much slower than assembly, and how they managed to put it together quicker than earlier Sonic games.
|
|
#
?
Sep 13, 2014 10:55
|
|
only
|
ratbert90 posted:
Neither of those lines are correct, they need human-readable names (Yes, including for the addresses, unless the address is literally the only thing in the define) Edit: Actually, where did you get that version? Freescale's 3.0.35 is currently: No I'm literally retarded, I have the same code as you do. Harik fucked around with this message at 14:44 on Sep 13, 2014 |
|
#
?
Sep 13, 2014 14:24
|
|
|
Blotto Skorzany posted:On a somewhat less serious note, NBA Jam was written in assembly (in 1993!), as were most games before about 1990.
|
|
#
?
Sep 13, 2014 15:56
|
|
|
Harik posted:Neither of those lines are correct, they need human-readable names (Yes, including for the addresses, unless the address is literally the only thing in the define) I want to move to kernel 3.10, but the project is almost done and .dts files scare me. 
|
|
#
?
Sep 13, 2014 16:47
|
|
|
qntm posted:There was a brief time when I served my website with a Content-Type: application/xhtml+xml HTTP header, which forces browsers to treat the payload as strict, correct XML or else display a parsing error. But even though my website has perfect XHTML markup (even the comments, thanks to draconian input validation), several moderately obscure browsers refused to render it as a web page, and I had to revert it. If someone with Konqueror complains that your website isn't rendering for them, gently caress em.
|
|
#
?
Sep 14, 2014 00:58
|
|
|
Volmarias posted:If someone with Konqueror complains that your website isn't rendering for them, gently caress em. The point of standards is that they make the web work for everybody, even people using marginal browsers. So I hope what you're saying is "if somebody with a broken browser like Konqueror or Internet Explorer complains that your website isn't rendering for them, gently caress em."
|
|
#
?
Sep 14, 2014 04:25
|
|
|
xtal posted:The point of standards is that they make the web work for everybody, even people using marginal browsers. So I hope what you're saying is "if somebody with a broken browser like Konqueror or Internet Explorer complains that your website isn't rendering for them, gently caress em." Yes, marginal browsers that are correct get a web. Major incorrect browsers also get a web, because the point of authoring a page is to reach people, not to satisfy the lust of a validator.
|
|
#
?
Sep 14, 2014 04:28
|
|
|
Not when it's a considerable amount of effort to reach them. For high-profile sites like Facebook, yeah, you still need to support IE6 in tyool2014. For anybody else, writing an IE6-compliant site is a big loving pain, and we shouldn't have to do it.
|
|
#
?
Sep 14, 2014 06:56
|
|
|
Yeah, I agree; I don't think IE6 is a major browser any longer. (I don't think we even support fully it on the main FB site.) I don't think supporting broken large-user-base browsers is a moral imperative, I just think it tends to happen a lot.
|
|
#
?
Sep 14, 2014 07:32
|
|
|
SupSuper posted:Given the sad state of C compilers those days, it's not surprising. Hand-coded assembly was still a major component of games up to around ~95, until the scale of games and teams started way outweighing any performance benefits. Yeah it's important to understand that codegen and optimization in C compilers was really, really awful for a long time. Using C removed a bit of tedium, but most of the time it simply wasn't worth the performance hit and memory overhead. With practice and an assembler with a decent macro system you can write fairly high level code. On the oldest machines the hardware had a lot of useful intelligence built into support chips which effectively gave you powerful "APIs" for graphics and sound, and later on PCs you had access to useful libraries and utility routines via BIOS and OS APIs that achieved the same thing.
|
|
#
?
Sep 14, 2014 15:50
|
|
|
Subjunctive posted:Yeah, I agree; I don't think IE6 is a major browser any longer. (I don't think we even support fully it on the main FB site.) I don't think supporting broken large-user-base browsers is a moral imperative, I just think it tends to happen a lot. ie6 is tricky for us because we do have a big(ger) presence in places like china where usage is pegged at ~10-15 percent of the population. how good the feature has to work in ie6 is heavily dependent on if it will be internationalized to markets where its a serious proportion.
|
|
#
?
Sep 14, 2014 19:43
|
|
|
FamDav posted:ie6 is tricky for us because we do have a big(ger) presence in places like china where usage is pegged at ~10-15 percent of the population. how good the feature has to work in ie6 is heavily dependent on if it will be internationalized to markets where its a serious proportion. Yeah, if we worked in China the calculus might be different. Though honestly supporting IE6 would be so much work that I could see arguing for just not addressing that submarket.
|
|
#
?
Sep 14, 2014 20:38
|
|
|
Subjunctive posted:Yeah, if we worked in China the calculus might be different. Though honestly supporting IE6 would be so much work that I could see arguing for just not addressing that submarket. we've managed to make it saner over the past several years by adopting (and enforcing) a style guide w/ the necessary libraries to make it all happen. before that it was very much so a free-for-all, developers having to reinvent and retest what should be common components, and our site design being the worse for it.
|
|
#
?
Sep 14, 2014 21:14
|
|
|
Edison was a dick posted:I remember reading an article about how Sonic Pinball was the first Sonic game to be written in C rather than assembly, about how everyone thought it was a crazy idea at the time, how they don't think it was much slower than assembly, and how they managed to put it together quicker than earlier Sonic games. Yeah, it was one of the first console games in general to use C (I won't go as far to say THE first, but it's up there). That said, I don't think the game really made a great argument for its usage.
|
|
#
?
Sep 15, 2014 00:50
|
|
|
qntm posted:There was a brief time when I served my website with a Content-Type: application/xhtml+xml HTTP header, which forces browsers to treat the payload as strict, correct XML or else display a parsing error. But even though my website has perfect XHTML markup (even the comments, thanks to draconian input validation), several moderately obscure browsers refused to render it as a web page, and I had to revert it. I remember trying to get that to work. You had to return a different content type based on the user agent the browser was reporting because not all of them liked application/xhtml+xml, but at the time spoofing UAs was so common because of all of the lovely old code that would tell, say FireFox users, to 'upgrade' to IE 5.5 or Netscape 4(?) that it wound up being basically impossible to support XHTML.
|
|
#
?
Sep 15, 2014 17:55
|
|
|
I can't decide whether to leave this as is or to handle surrogate pairs properly  Java code:
|
|
#
?
Sep 16, 2014 17:44
|
|
|
1337JiveTurkey posted:I can't decide whether to leave this as is or to handle surrogate pairs properly What is this even trying to do? Also, "௫"
|
|
#
?
Sep 16, 2014 18:20
|
|
|
b0lt posted:What is this even trying to do? I guess it's massaging a string that's supposed to be a number so that it's always a number (for certain values of number) when fed to a downstream third party system. Since the string is ultimately just an opaque identifier anyhow, so long as everything maps the ID consistently it works (for certain values of works). Edit: Call it an industrial grade injective mapping 1337JiveTurkey fucked around with this message at 18:49 on Sep 16, 2014 |
|
#
?
Sep 16, 2014 18:41
|
|
|
So if you're working on a legacy database suffering from the results of no input validation, duplicate keys (6 that mean "It's a file" in a type table, but 3 File and 3 FILE...) and keys with no description, keys used in the document table that don't match anything in the type table, and descriptions in the type table with no key associated, is that a database horror? Is it a bigger horror that the numbers you search for files with should be "year-2LetterCode-6digitNumber-+4 characters maybe" (1234-AB-123456-XXXX) but with no input validation many times the year, code, or both are missing, so you can't split it into columns and do proper index searches and have to just do a LIKE '%foo%' with a drat scan? Oh and you can't edit the database in any way anyhow, so you're stuck with unindexed scans. Thankfully it is used maybe a few times a month, so it can afford to be slow - it's only 1.8 million rows, so those ugly scans don't even take a second, but my god, I'd like to think people knew what input validation and schemas were. The other scary thing is it's still around because it's tracking non-digitized, printed documents! The cherry on top is the location they're stored at is often $PERSON's desk or $PERSON's office, not SECURE_FILING_CABINET or REPOSITORY, and more than one time said document was found propping up a table leg because the floor settled. 
|
|
#
?
Sep 16, 2014 19:45
|
|
|
b0lt posted:What is this even trying to do? code:
|
|
#
?
Sep 17, 2014 13:57
|
|
|
Oh boy. I get an IM from a coworker:IM posted:now THAT... is a SQL statement I naturally laugh and ask him where he found that thing: IM posted:I had to write it from scratch >_< 
|
|
#
?
Sep 19, 2014 21:35
|
|
|
|
| # ? Jun 7, 2024 07:29 |
|
|
Ruby code:
|
|
#
?
Sep 20, 2014 00:29
|
|
