links to other domains then www.librarything.com and anonymous users redirect to /?highload=1
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.
I noticed the significance / implications / consequences of this report yesterday.
English users can post LT urls anywhere without asking friends / people to click on a link twice. The relevant contant for that link is displayd (or a cached variant).
However if another subdomain is specified "/?highload=1" will ask users to sign in first. Such strategy is not known from other multilanguage sites. It would be fair to have the same behaviour for all LT languages i.e. for English and all other LT languages.
Thanks for any help in advance! Best regards Reinhardt.
P.S. I noticed this in connection to posts on Facebook using the www.librarything.de subdomain.
I just wish the forced login would preserve the URL you were originally going to and forward you there after login. Plenty of sites do it; it's not exactly rocket science.
By the way, but that doesn't address the principle of the matter, I noticed that if you use the "Back" button of the browser, and then try the link again, usually you suddenly don't have to log into the other language site.
Well, yes and no. Most of the users use the English site. I think it's pretty appropriate that it get more resources. If this is a way that Tim found to avoid paying for more resources for what is, overwhelmingly, an English site, I think that's fine.
> 4: Right, like I said, it doesn't address the principle of the matter.
> 5: Hm, yeah, I can see your point. For me it's just a slight frustration when I have to follow some link in talk. But I do see gangleri's point too.
#6 by BarkingMatt> Oh, I see his point, too. I just think he's a very atypical user. And really, what's the big downside here? You login to the site and it keeps you logged in for a year. The only real way you run into it is if you're a user who uses lots of different non-.com sites frequently, and are on a locked down machine where you can't stay logged in. So that's gangleri and... who else? :D
It's possible the highload thing isn't necessary anymore and Tim just left it in. If so, I'm all for removing it. But I'm going to start with the assumption that there's a reason for it and work towards "are enough people affected to overcome that reason?"
A basic question: Does LT need new users? Should references to LT be in English only? As it is now refferences in other languages are useless. This mighte be acceptible for Danes, Dutch, Swedish, Finish and other people who are learning English as second language but not for older people (some being authors).
One year ago I received a vary angry e-mail: "How much get you paide to atract people to a paid site?" Being a registred user I did not understand where this thoughts originate. My German Wikipedian girlfriend never signed in and meeting her last month she said she would never sign in and contribute with anythink. This is one can kind of marketing. It is not up to me to comment on this. I addressed the issue ind it is up to LT to decide on it.
I really don't see how making these people log into the other language sites once a year will make any kind of noticeable dent in LT's appeal to non-English users.
I'm also baffled by what you're trying to tell us with your German wikipedia girlfriend (!) example.
>9 I realy ask myself why I contributed five years mainly in English, stupid me.
Ah, now I really see what you mean. It's that non-registered users - people visiting the site for a look around - can only easily gain access to the .com site, not the one in their language. (I was first just thinking about registered users who prefer other languages - sorry). Yes, that shouldn't be.
>12 Please let stuff decide if it is a bug or not. It is more like that you and I have different POV's (point of views).
Not a bug.
A bug is something that doesn't work as it was intended to, not something that you'd prefer to have work differently than it was intended to.
It's at least potentially a bug, brightcopy's conjectures aside. As I recall highload is intended to keep web crawlers from using the site when it's under high load. What that has to do with this situation is a mystery and may well be a bug.
Right, so maybe it isn't a bug as such - big deal. That's why I went on to post this in RSI. The real issue is that's something isn't working as one would expect. The site can improve if we solve this - isn't that the main thing?
Why can't it stay opened until someone who really knows determines if it's a bug or not?
It can. I'm certainly not going to engage in an edit war.
Just like if someone files a bug saying "LibraryThing" should be written "Librarything", I'll close it but won't re-close it if someone reopens it. I just follow all the guidelines Tim has given us when he gave us the ability to close bugs. i.e. NONE! ;)
(Not that I fell this report is as trivial as the example.)
I've just been sent to the .nl site by a link, and got redirected to the highload page. It's 5:39am BST, which is middle of the night in the US. There are 120 members online.
Perhaps the real bug is either that something's causing an abnormal load on the servers, or the calculation for the /highload page is wrong?
Check out what I say here:
Thanks for the statement Tim!
What about some exceprions as:
http://ru.librarything.com/zeitgeist/language/rus, http://www.librarything.de/zeitgeist/language/ger , http://epo.librarything.com/zeitgeist/language/epo etc. ?
All links starting there can behave as implemented today.
P.S. Faulty links as http://www.librarything.de/zeitgeist/language/de (generated manually) generate "zero" statistic. Garbadge in garbadge out.
Katalogisierte Bücher: 0
Gruppen: 0 (Gruppen anzeigen)
Die größten Bibliotheken
Die fleißigsten Kritiker
Mit mindestens 10 Büchern
If stop those pages from requiring a login, web spiders would request EVERY PAGE those pages link to. That's exactly Tim's point for preventing it.
ETA: Plus, all it would in the spider's search result is a link to the language page. I don't really see how that's a good search result that will be of interest to non-LT users.
>23 I understand both Tim's argumentation and the problem about the traffic load caused by bots / spiders.
However the textual content of http://www.librarything.es/zeitgeist/language/spa gives a much better language overview then http://www.librarything.es/?highload=1 (spiders will try to access these pages also).
The compromise could be to have some (variants of) expressive (in German "aussagekräftig" http://www.dict.cc/german-english/aussagekr%C3%A4ftig.html ) pages with minimal clickable links; i.e. "textual" information might increase the interest to register at LT.
One big difference is that on http://www.librarything.es/?highload=1 there are around 100 unique links. On http://www.librarything.es/zeitgeist/language/spa there are around 300. That's 200 hits more. For EVERY language site. For EVERY spider. For EVERY time it re-crawls the site.
Plus, all of this requires that somewhere else there must be a link visible to the spider to http://www.librarything.es/zeitgeist/language/spa (and all the other language sites).
In my opinion, it all sounds not worth the trouble.
>25 "On http://www.librarything.es/zeitgeist/language/spa there are around 300." (more)
On a textual implementation there are ~ 54 links for the LT subdomain languages and a few askiing to register. I used "textual" which means not linkified.
He wants a separate, plaintext version of the Zeitgeist, with no or fewer links to crawl, to be made available to spiders and non-logged-in users. I'm not sure why, but he sometimes is concerned with random google searchers ending up on arbitrary places on LT and what they will think of what they see.
#28 by lorax> I'm not sure why, but he sometimes is concerned with random google searchers ending up on arbitrary places on LT and what they will think of what they see.
The thing is, that'd actually make them MORE likely to land on an arbitrary place on LT. An arbitrary place where the links didn't work!
This topic is part of LibraryThing's in-talk bug tracking.
Join or watch Bug Collectors to get "Bug Tracking" under "The World" in Talk all the time.
Category: Non-English LibraryThing
Assigned to all
Reported by gangleri
Status: Closed by staff
Sep 25, 2012, 10:09am
1 years since last change
This topic is not marked as primarily about any work, author or other topic.