Join LibraryThing to post.
This topic is currently marked as "dormant"—the last message is more than 90 days old. You can revive it by posting a reply.
Having waited in vain several months for the problems with searching to be fixed, I now leave LibraryThing. I just changed my profile text to this:
"I have removed all my books, because LibraryThing is only for book in English. That becomes evident when one tries to search for books where the titles or the author's name contains non-ASCII characters. That is simply not possible. In spite of many complaints over months, the administrator refuses to correct that. Therefore I now definitely stop using LibraryThing."
The Google option added to the search page would have solved a lot of that - but it had to be taken down again because some of the results came from libraries defined as private.
Yeah, but not all of it. Character support needs to happen thru out the entire system. I imagine it's a big job and there are more important features the majority wants/needs.
It is a big job, and something we're aware of. We've made a number of internal changes to support this, but because they're so deeply embedded in the system, we can't just turn them on. We want and plan to improve our handling of non-ascii character sets. I feel badly we haven't been able to do it more quickly, but we are working on it.
This message has been flagged by multiple users and is no longer displayed (show)
"the administrator refuses to correct that"
This is what Tim gets for bending over backwards for the past year or more to prioritize all these damn foreign language versions ... until L.T. is searchable in every alphabet (all the way down to archaic Klingon glyphs) you'll have somebody with their panties in a knot over some minutia of their particular local scratchings not popping up top of the list.
I get so sick of the Wogs having hissy fits about crap like this, while there are dozens of way higher priority issues that need fixing!
"somebody with their panties in a knot over some minutia of their particular local scratchings not popping up top of the list."
He had his Esperanto library in here (and Bertilo W. is a bit of a household name in the Esperanto world). Esperanto has a number of accented letters, that are not part of the usual Western European character set (and you can totally forget about standard Ascii, which should be outlawed). There are hiccups with German and French, but I can imagine the pain for someone wanting to search his Esperanto library...
"I get so sick of ..."
Well, I am not going to comment on that.
This message has been flagged by multiple users and is no longer displayed (show)
"in the Esperanto world"
Let me guess, that's a left-turn past the Paisley Pony world?
Here we go again.
I don't see a personal attack in #5. Quite a broad attack on all those wanting non-English character support, actually. Or I guess it is. If you follow the Wogs link, you find an article discussing the arrogant attitude of Americans towards foreigners. :-)
Why not just leave it to speak for itself? Or bring it to Tim's attention if you think it violates the TOS?
I never thought to flag that post (though I do think the post was rude and unnecessary), but the flagging guidelines are a bit strange. The TOS says "be polite", but abuse doesn't include all TOS violations. I can't imagine why it would be okay to write a random abusive post attacking various ethnic groups or religions; such a post would certainly be flagged despite not fitting into one of the three categories. Maybe the guidelines should be changed?
#8, myshelves: The link wasn't there yet when I wrote my message, so I had to take it at face value. And left my opinion on it (no, no flag from me, I think we are all able to read).
#7, btripp: Maybe, but what I meant, is that he could have been a multiplier, and now we've lost him. Maybe it's only a special interest group, but it's worldwide :-). I knew his name *before* I knew of LibraryThing.
Maybe it's only a special interest group, but it's worldwide
Yes, I agree, but one must realize that LT must prioritize its operations, and the interests of a small special interest group are unlikely to be at the top of the list (here or anywhere). And, as John says, (see #4) the requested fix isn't so easy to do!
I do think it's a shame that this conversation started to degenerate into name-calling. I hope that that has stopped now.
It wasn't me! Rude post, yes. But there wasn't a personal attack, which I've been told by Tom is the guideline for flagging.
Chill out a bit!! LT just lost a user, a fellow bibliophile, a companion (in the broad sense of the term).
I, for one, am disheartened by the bickering in this thread. Tim has said that the software we're running on is in a constant beta state (or something close to that, anyway). LT is still evolving. It’s well beyond having developed opposable thumbs and tying simple knots, but hasn’t mastered many languages yet. Frankly, I’ll be thrilled when I can read everything in Nordic Runes, but there are bigger priorities, and I’ll wait patiently.
Having been a programmer and bug-fixer ("never my own code", he brags) for over 25 years, I feel I can say with some authority that adding alternate languages, especially when it comes to adopting multi-byte character sets, is a daunting, time-consuming task. I fear you've been spoiled with years of downloading programs that set themselves up using background scripts, and have little understanding of what it takes to make that happen.
From personal experience, I've made a simple 5-line fix for an SPT (Stupid Programmer Trick) in a fairly high-profile program, and it took almost 2 years to get it into Production. Politics, Bureaucracy, SMT (Stupid Management Tricks) caused this senseless delay. Tim and company don’t work that way. They have been as responsive as time and resources can allow them to be. They are users as well as the maintainers of the system and the code. It truly is a matter of time/resources relative to benefit-received.
Not to argue, but re: outlawing the standard ASCII character set – get real. Perhaps it could be replaced with a 128-bit character set that could ~possibly~ meet the requirements of all Earth-based languages (okay, a 256-bit set, perhaps), but it won’t be any time soon. There’s too much invested (outside of LT) to make that a near-time reality.
I am surprised by the heat of the original poster. LT is based in the US and most of its audience is English speaking. Trying to make something work globally is a HUGE task for something like Google. I can't imagine how much more difficult it is for a little Web 2.0 startup. It can't be all things for all people. Not with three people on staff!
Message 14: Morphidae -- I agree with you! For all we know, this user is a nut case. After all, if I were disappointed in a website, I would just quietly crawl away. I wouldn't post a topic in a forum about it. I am astounded that in less than two years, Tim has developed the ability to handle all the languages LT does. Is it fair to criticize an English-language site that doesn't handle well every single obscure language in the world, especially weird made-up ones?
If the website had a place for suggesting improvements, I would make it known what improvements I thought were needed on my way out. I'd think they would appreciate that, because surely it's useful to know why people are leaving.
I guess I don't see it as a 'small special interest group'. Nor is it particularly a 'small problem'. This problem extends to all character sets beyond, approximately, 'western European languages' and may affect 'core functionality' from a user point of view, not just search.
None of Chinese or Japanese character author names generate an author page at all, the system doesn't recognize the characters. The titles are uncombinable because they don't share an ISBN and they have no author page, much less an author page in common. That means there is no LT social networking for books in these character sets and for some character sets all entries are manual, which speaks to a fair bit of dedication on the part of the users to get a large list of books in LT.
(LOL! You should have seen the tears of gratitude when server crashes stopped corrupting Japanese characters in titles and author strings. I don't know that I would have had the fortitude to correct them all only to have them corrupted again on the next crash.)
I trust the problem will be solved in the long run, and consider LT well worth the wait, but not everyone has the patience to wait for it.
Given that Esperanto is a smaller community with rarer books and a potentially higher social networking need, I think bertilow's loss is a great loss to the LT community. I hope someone will invite bertilow to take a second look at LT someday when these problems have been solved.
I think what people find offensive is In spite of many complaints over months, the administrator refuses to correct that.
We all know that "the administrator" has a name, and that Tim Spalding is very responsive to user concerns. We also know that he can't do everything yesterday.
Of course, now that someone couldn't resist adding that 4th flag, and history has been rewritten, we can all forget about the other offensive post, and concentrate our criticism on people who want their particular local scratchings to show on LT. (Sarcasm.)
> surely it's useful to know why people are leaving
> most of its audience is English speaking
But LT offers import from so many foreing language sources, and already there the non-ASCII characters often don't come through correctly.
> every single obscure language in the world, especially weird made-up ones
Ok, I guess you _could_ apply
that to French, German, Spanish, Portuguese, Danish.. ;-)
It hasn't got to do with letting the users translate the interface; I have books in those languages (ok, probably only a dictionary in Portuguese) that I would like to be searchable, combineable etc. - and what's more than that: I'd like to be able to search for authors and title in those languages.
BTRIPP's frustration is not so long away from bertilow's, though: both are dishearted that the LT team isn't working on the issues they consider most important.
(Although I do think that BTRIPP could stop blaming it on LT being translated to foreign languages as he could know by now that it was the users who did it, once translating was made possible.)
Me? I think search should work much better (English and other language titles). Google search looked promising, so I hope it will be back. But it won't solve the wrong characters at import.
But mainly I think Tim, Abby and John put so much into LT that it is important that they keep liking it and can work on features that _they_ consider worthwile.
It was wrong to flag it, but that it still can be read under 'show' is a clever solution.
For Japanese - what's wrong with using Romaji? "spelling corrected: oops"
So far I've catalogued books in Norwegien, Japanese (okay, it's only a dicitonary published by an Italian, but it counts...), and Arabic (transliterated into English). None of which I had problems with.
Hunting down the information for the Norge book was a bit of a pain and when I entered it, there was only me. Now there are two others. (Apparently they found it useful without the umlaut.)
It is unfortunate that we can't use all the accent marks and umlauts and Kanji and bantu script right now. From my younger days, I can't recall seeing any accent marks in any card catalog when I was looking for foriegn language books. Why should LT be any different?
The version of rōmaji that most English speakers are familiar with uses macrons for long vowels. So you are somewhat better off, since only some letters are ignored, but the problem still isn't solved. There are other schemes (oo or ou or oh for ō), but you'd have to remember to use one consistently.
The romanization for other scripts is even more dependent on diacritics. Again, there are alternatives used in ASCII-only chat rooms or for text messaging, but they aren't what one sees in textbooks or library catalogs.
Esperanto has an unambiguous scheme using x in place of accents. So the problem isn't that it's impossible but a perceived lack of commitment to the way the user wants to organize their library.
Most English speakers? I'd have to disagree.
Now that anime and manga have gained quite wide fan-base in West, hardly anyone uses ō. In fact, I don't remember when I last saw somebody using it; my Japanese teachers certainly didn't.
It's standard to write either shōnen or shounen, but ō definitely doesn't equal oo in all cases. There's no such word as しょおねん, but on the other hand, ローマジ is an actual word. ō can be either ou or oo, depending on the word. It's clearer to just avoid the usage of ā, ē, ī, ō and ū - though admittedly, hardly anyone actually writes rōmaji as roomaji: it's far more common to see somebody misspell it as romanji.
oh isn't even Japanese, so that one is definitely better left unused and forgotten. It originated from the passports, and hardly anyone actually uses it, except occasionally (Yu-Gi-Oh).
13> "Not to argue, but re: outlawing the standard ASCII character set – get real. Perhaps it could be replaced with a 128-bit character set that could ~possibly~ meet the requirements of all Earth-based languages (okay, a 256-bit set, perhaps), but it won’t be any time soon. There’s too much invested (outside of LT) to make that a near-time reality."
It's called Unicode.
It's technically possible for us to support pretty much every script in the world (using Unicode/UTF-8; Joel Spolsky gives an excellent an overview of character set issues here).
We aim to do this in a more thorough way, and I think it's a completely reasonable expectation that we do. The only reason it isn't done yet is that the system was set up before we fully understood how character set issues related to our database. And now that there are 11 million books in the db, it's more difficult to fix than if it had been set up to handle internationalization off the bat. It's the whole fixing-an-airplane-while-in-flight thing: Harder than when you're on the ground. I think I'll set up a keyboard shortcut to print out this phrase, but... we're working on it.
Tim would agree, I think, but he's speaking at a conference in Denmark. Where they use a character set that doesn't fit into 7-bit ASCII.
Yeah, I basically agree. It's complicated by the fact that MARC records come in such different character sets. (I just came from Denmark, a nation of five million people with its very own version of MARC.) If you want some more excuses, do some searching on the Library of Congress website; they screw of character sets quite thoroughly too.
The search issue is tied up in the fact that searching in general is not good enough on LT. When we rewrite the system, we'll rewrite the character set issues pertaining to it. Adding the Google search was a step in that direction—to provide something to stand on while we ripped up the flooring.
Lastly, our first priority is to get LibraryThing working for the languages with LibraryThing domain extensions. Except for Greek, they're all Latin languages. I'd like to expand that, but the first priority is to get all the Latin-alphabet languages working, and to localize the site a bit more—with Z pages by nationality, notice of what groups are in your language, etc.
I’m sorry such a famous Esperantist has left. Will you announce when LT is Esperanto-ready (or any other language-ready, for that matter)?
This group does not accept members.
This topic is not marked as primarily about any work, author or other topic.