?

Log in

No account? Create an account

S2 style comment viewing broken

« previous entry | next entry »
Mar. 14th, 2015 | 11:43 am

Journal pages that include replies to comments are broken in (some) S2 styles right now; if a page ends unexpectedly, with a message saying

Error running style: Died in S2::run_code running EntryPage::print(): Can't use an undefined value as an ARRAY reference at (eval 17601)[/home/lj/src/s2/S2.pm:203] line 2631, line 13855.

or so, you're affected. This has already been reported to LJ (many, MANY times, in fact); until they unbreak S2 styles, you can work around the problem by appending ?format=light to an entry's URL. (Be sure to do so before the fragment identifier if there is any, of course.)

EDIT, 2015-03-16: this doesn't affect all S2 themes; themes based on the Component layout are affected, others may or may not be.

EDIT, 2015-03-24: I didn't pay attention to when it happened exactly, but apparently the issue's been fixed, and everything's back to normal.

Link | Leave a comment | Share

Comments {20}

The Mystery of the Supranational Rabbit

(no subject)

from: porsupah
date: Mar. 14th, 2015 01:09 pm (UTC)
Link

They do seem to adopt a "WON'T FIX" attitude on occasion, as with that failure to post a new update, as I encountered recently. Their reply?

"Thank you for your report. LiveJournal has blocked the ability to post links to Deadline.com due to actions taken by Deadline.com's legal department. Removing that link should allow the rest of the content to post normally."

Petty at best, but ridiculous considering this must've been years ago, as I encountered the same problem previously some time back. O.o; (And so damned annoyingly implemented, too, given the error doesn't give any detail as to what's wrong - I recalled it was some site being blacklisted, so I went deleting links one by one, attempting to post after each change, and turned the "offender" into a shortlink)

Reply | Thread

Schneelocke

(no subject)

from: schnee
date: Mar. 14th, 2015 04:39 pm (UTC)
Link

Aye — I wrote about that previously. I'm not sure what's going on with between LJ and that site; I'd speculate that they got a nastygram because some users deep-linked to content that Deadline.com was unhappy having deep links to, and as a result they decided to disallow linking to the site altogether — more for their own business benefit than the users', of course.

If this is indeed the case, it's understandable, yet still annoying. And of course an error message that actually indicates what the problem is and helps you fix it should be a no-brainer.

This new error should be fixed, I think, given that it's not about content (especially third-party content); it's just a bug in LJ's S2 style code. Chances are someone deployed some new code on the production cluster without testing it properly. (Or it was tested, but the tests themselves are inadequate. Or the tests are inadequate, but can't possibly cover everything because S2 is so baroque. The possibilities are endless. :))

Reply | Parent | Thread

ungulata

(no subject)

from: ungulata
date: Mar. 14th, 2015 03:24 pm (UTC)
Link

Thanks for the heads-up. ^_^ Should I run into any trouble, I'll know where to go to reread the solution. So far, my policy of no JS on LJ continues to remove most of the annoyances and simplify navigation for me. I can deal with the stark reading page and the 10 second 'switch to ?nojs=1' delay.

Reply | Thread

Schneelocke

(no subject)

from: schnee
date: Mar. 14th, 2015 04:40 pm (UTC)
Link

You're welcome! In addition to ?format=light there's also ?style=mine, BTW, but unfortunately that doesn't work when your own style's one of those that are broken. :P

Reply | Parent | Thread

ungulata

(no subject)

from: ungulata
date: Mar. 15th, 2015 12:38 pm (UTC)
Link

Oh ho! Got that bug reading this entry today! Blocking cookies and JS did not fix it, only your solution worked! Thanks for the tip. ^_^

Have you noticed just how more restrictive the internet gets with each passing day, from functionality to region blocking to 3rd party content filtering?

Reply | Parent | Thread

Schneelocke

(no subject)

from: schnee
date: Mar. 15th, 2015 01:41 pm (UTC)
Link

Yeah, it's a server-side issue. LJ's deployed code is broken and not returning the right data; the only thing you can do client-side is avoid triggering the buggy code (e.g. by avoiding S2 entirely).

And yes, that's been an ongoing trend. Cory Doctorow called this the "Coming War on General Purpose Computing"; Schneier's concept of feudal security is related as well. It's also implicit in e.g. Mozilla's manifesto, which is fairly explicitely intended to counter this development.

I personally think that the commoditization of the Internet is to blame, in multiple ways. People are more likely to access the Internet on portable devices these days, and we have long since accepted that many devices, unlike "real" computers, are not open: that we do not have the same level of intimate access and cannot do whatever we want on and with them. (This is why I am suspicious of both tablets and modern game consoles, for instance.) There's a variety of justifications given for this, from convenience to security (lay users are notoriously bad at keeping their machines patched and secured), though in actuality it's more down to control and the ability to monetize users and their data.

At the same time, commoditization also brought large numbers of people to the Internet who had no previous exposure to its culture, to its ideals of openness, and in fact to general-purpose computing as such. When relatives of mine (in their 70s) bought a computer for the first time in the mid- to late 2000s, they only wanted to be able to do such things as exchange emails, call the kids on Skype, perhaps shop online or compose letters in a word processor or so. The ability to install whatever program they wished and run arbitrary code was meaningless to them; ask the current generation of Internet users if they know Barlow's Declaration of the Independence of Cyberspace, and most likely you'll just get blank looks all around.

Of course restrictions and closedness eventually come back to bite you in the ass, even if the golden cage appears enticing at first. Got an iPad, want to play a Flash game? You're out of luck, on multiple fronts in fact: Apple won't allow Flash on its devices, and Flash is a proprietary format with a proprietary player anyway, so you're dependent on Adobe to provide a current player even on supported platforms. (Just ask any Linux user how well that's working out.)

People will eventually learn this, I believe. Ebooks that come with DRM restrictions aren't yours. Software you bought that is "in the cloud" (again, Adobe comes to mind) isn't yours. Devices that are controlled by third parties (Apple, Google, Facebook, whatever) aren't yours. You're not free to do what you want with these, and eventually this will hinder you when you try to do things that the manufacturer of the device didn't contemplate or (worse) did contemplate but decided would interfere with their bottom line.

Spying on users is another issue along these lines. The kerfluffle re: smart TVs and the various ways they collect data and surreptiously send it back to third parties is a good example; even relatives of mine who're not technically inclined are spooked by this (particularly those who bought smart TVs in good faith).

The main problem I see with all this is that it's easy to hide it all from peoples' view: restrictions, spying, lack of control etc., it's often all very difficult to see unless and until it's too late.

(cont'd)

Reply | Parent | Thread

Schneelocke

(no subject)

from: schnee
date: Mar. 15th, 2015 01:41 pm (UTC)
Link

(cont'd)

Sometimes trade-offs are worth making, of course. For example, years ago I thought I'd never get a game on Steam; last year I did, and ended up getting hundreds of hours of fun out of it, for a low two-digit amount of money. I may not own the game in any meaningful sense (despite owing a physical, boxed copy, and despite it being a single player-only game, I cannot even run it without an Internet connection and working access to Steam's servers), but I still feel I got good value for my money, for now. But this still depends on the user having the facts on the table so that they can make an informed decision in the first place.

At other times they're not. I've never bought an Ebook, for instance, and I don't foresee it happening anytime soon, unless I get an unrestricted PDF. (Even then, I'd still want an unrestricted, general-computing reader device.)

Anyhow — as for the Internet again specifically, one thing that's different is that large/multinational corporations exert much more influence nowadays than they used to in e.g. Barlow's days. Google in particular is notoriously pervasive; even if you eschew its services and block such things as Google Ads, Google Analytics and so on, large numbers of sites will break without access to its AJAX API. Your browser likely uses Google for blocking phishing/scam sites. And have you tried watching videos anywhere but on Youtube? Except for the rare case where somebody opted to upload something to Vimeo or Dailymotion (instead or in addition to Youtube), you simply have no choice.

This gives companies such as Google vast amounts of power, and Google in particular's trying to go further. Google DNS, Google Chrome, Android, the list is endless; Google is also active behind the scenes, e.g. with SPDY (now HTTP/2.0) and so on.

Not all of this is a bad thing. And even where it is I don't want to single out Google; my point is just that the web of the early to mid-90s, which consisted of a diverse array of personal homepages and non-profit sites, has been replaced by a much more concentrated web where much power rests with specific companies. And those companies will exert control when it benefits them, or when they get pressured by third parties to do so (such as with Youtube and its region-blocking, content-identification, and streamlined user suspension independent of any legal process where the accused might, if nothing else, be heard at the very least).

All this said, it helps to remember that in the grand scheme of things, the Internet is not important. Humanity has survived without it for long, and we shouldn't take it TOO seriously.

Sorry for the ramble, BTW.

Reply | Parent | Thread

ungulata

(no subject)

from: ungulata
date: Mar. 15th, 2015 04:47 pm (UTC)
Link

That was a very interesting ramble, well worth two replies to fit it all in. So much truth and so much that I've forgotten.

In the late nineties, with all those personal homepages and not-for-profit specialized sites, it seemed like soon you would have giant webrings on every topic in great detail, that you could cross check and cross reference endlessly. All knowledge and every taboo would meet in a grand crunch of new openness and enlightenment. Instead, we have Wikipedia and little else.

Wikipedia tends to be short and to the point. Say you want an exhaustive database of all the Bambi merch and ephemera made between 1928 and 1975? Not going to happen. How about a guide to building chip controlled robotic limbs that use alcohol activated memory metals? Not very likely. Research and reprints? Oh no no, there's money to be made printing those.

Buy all the journals. We're bottling the genie.

Reply | Parent | Thread

Schneelocke

(no subject)

from: schnee
date: Mar. 16th, 2015 11:30 pm (UTC)
Link

Thanks.

Yeah, it's a shame that webrings died out again so quickly — as did the hand-curated lists of useful/relevant links that preceded the first search engines on the very early web. Wikipedia is useful, but has its problems, systemic bias being one of the biggest ones IMNSHO.

Wikipedia's focus on creating an encyclopedia rather than a free-for-all site isn't that much of an issue for me, though many Wikipedians have an altogether too narrow view of what an encyclopedia may contain. Traditional encyclopedias were selective for a variety of reasons, including physical dimensions and printing costs; Wikipedia is not paper and can afford to not be. Many people also seem to have this idea that in order to be accepted by the more conservative parts of society, academia etc., Wikipedia must strive to be an online Brockhaus or Britannica or whatever, a site where Pokémon and Ponies have little place; but I think this is misguided, both because those naysayers are only a small share of Wikipedia's audience, and because you're never gonna win over (many of) those people anyway.

There's sister projects, though. A guide to building chip-controlled robotic limbs that use alcohol-activated memory metals might be welcome on Wikiversity or Wikibooks, perhaps; and exhaustive databases of ephemera could go on Wikia (though that's a for-profit platform not affiliated with the Wikimedia Foundation, of course).

Ponies have their wikis on Wikia: mlp., mlpfanart., mylittlebrony., and so on.

Some other sites have also outgrown Wikia and moved to their own hosting; I've occasionally edited a few that have. (Though in at least one case the Wikia site remained after the move, and was taken over by people not actually affiliated (much) with the community in question, with the result being that there's a "real" site and a lifeless animated husk on Wikia that keeps shambling along like a zombie.)

For research, there's sites like arxiv.org, too — with a fairly narrow focus, of course. In any case the professional publishing industry (scientific journals, that is) has some major changes ahead of it; the old system worked well and was necessary when journals had to be printed on paper and distributed to get knowledge into the hands of researchers, but these days they're an impediment as well. Peer review and journal editing are sorely needed, perhaps more so than ever, but closed journals that most people (including those whose taxes paid for the research) can't access will have to adapt or perish.

Buy all the journals. We're bottling the genie.

I'm not following.

Reply | Parent | Thread

ungulata

(no subject)

from: ungulata
date: Mar. 17th, 2015 03:02 am (UTC)
Link

That last line was actually two thoughts and probably should not have been placed on the same line. 'Buy all the journals' was a poke at the commercial interests making life difficult for the dissemination of scientific papers, and 'we're bottling the genie' is commentary on the increasingly restrictive nature of the "world wide web".

The MLP wikis are numerous and overlapping as well, but a welcome change from what you get on Wikipedia.

Reply | Parent | Thread

Schneelocke

(no subject)

from: schnee
date: Mar. 17th, 2015 08:29 pm (UTC)
Link

Ah, fair enough.

The main MLP wiki at mlp.wikia.com is pretty uptight as well as far as including information is concerned. They not just painstakingly keep out anything that's not official (this isn't a bad thing in and on itself), they also have a preference for namespace tidiness over usability, resulting in e.g. a lack of redirects and a tendency to lump minor topics into collections (e.g. "Creatures"), without giving everything an individual articles (in addition to having pages such as "Creatures" providing a general overview).

Minor points, but I'd do it differently.

Reply | Parent | Thread

Moth

(no subject)

from: moth_wingthane
date: Mar. 16th, 2015 10:44 pm (UTC)
Link

Mm, all of your entries are giving me this error when I try to read them :/

Reply | Thread

Schneelocke

(no subject)

from: schnee
date: Mar. 16th, 2015 11:15 pm (UTC)
Link

I've switched to a different theme for now (Tranquility II, the "Plain" variety) — this one's not affected by this bug.

Reply | Parent | Thread

ungulata

(no subject)

from: ungulata
date: Mar. 17th, 2015 03:04 am (UTC)
Link

Tranquility of theme for tranquility of mind. Et sur ce, bonne nuit. 8^)

Reply | Parent | Thread

Schneelocke

(no subject)

from: schnee
date: Mar. 17th, 2015 08:32 pm (UTC)
Link

It's a decent theme, though the main content container was a bit too small for my taste, with a specified width of just 800 pixels (including the sidebar). Fortunately that's easily remedied.

That said, though, I hope LJ will unbreak my old style soon. I liked that one.

Reply | Parent | Thread

Oni

(no subject)

from: onidemondog
date: Mar. 17th, 2015 06:41 pm (UTC)
Link

"Died"? How morbid! D:

Reply | Thread

Schneelocke

(no subject)

from: schnee
date: Mar. 17th, 2015 06:53 pm (UTC)
Link

It is rather — but then again it's not.

Reply | Parent | Thread

Oni

(no subject)

from: onidemondog
date: Mar. 17th, 2015 08:12 pm (UTC)
Link

That's pretty much like Greek to me. lol.

Reply | Parent | Thread

Schneelocke

(no subject)

from: schnee
date: Mar. 17th, 2015 08:34 pm (UTC)
Link

:) Well, the executive summary is that

  1. Livejournal is implemented in Perl, and
  2. die is a standard function in Perl, which is why the message says "Died in...".

That's all there is to it, really. The details of how die works aren't really of any concern unless you happen to be programming in Perl yourself. :)

Reply | Parent | Thread

Oni

(no subject)

from: onidemondog
date: Mar. 17th, 2015 06:42 pm (UTC)
Link

"Died"? How morbid! D:

Reply | Thread