At LinuxTag, Wikimedia Foundation chairman Jimmy Wales has announced plans for a co-operation with KDE to create “The knowledge-integrated desktop”. This will comprise two components 1) a Wikipedia web services API (applicable to any application / desktop environment) and 2) KDE APIs for easy integration of applications with Wikipedia / Wiktionary. This announcement follows news that the next release of the Amarok media player will incorporate Wikipedia lookup for instant access to band biographies.
I’d expect this type of expansion of open projects working together to continue.
With Wikimedia, Archive.org, and Wikipedia there are rich resources for content — maybe if all 3 were integrated? One can hope!
That means that many, many more people are going to see my far far too detailed page on Blue Monday now. Bollox.
I’ve had amaroK-1.3-beta1 on my suse 9.2 box for a while now (about 2 or 3 weeks?) and it already has an extra tab that when you click fetches info from wikipedia.
For example just now I was hearing the FF8 piano ost, clicked in ‘Wiki’ and was presented with the biography of Nobuo Uematsu, the composer (and the name that was in the artist mp3 tag).
Websites shouldn’t need an API to extract information from them.
While it sounds neat, from an academic point it’s somewhat horrifying. When I need accurate research I go to the library, not wikipedia. In today’s age there seems to be this implicit trust that if it’s on the internet, it must be true. That isn’t to say that there aren’t credible sources on the ‘net, but they’re far harder to verify. I’ve seen inaccurate information before in books and in WikiPedia, but it’s far rarer in books. Overall though, if they make this into a generic framework giving people the ability to integrate other sites, or RSS feeds it will be neat if only as a quick reference, but not a trusted one.
Do You mean that if it’s in library, then it must be true?
Because i doubt that libraries have some kind of license for truth. Every source of information should be checked, no matter what form it’s in.
As for idea to use wikipedia and similar sources in applications, i think it’s very good. I’m only not so sure about implementation, and i’m afraid about servers and bandwidth when more such apps will get more popularity.
I mean, if every application You use, connects to wikipedia to get some information (even if it will be cached later) it may be like DOS attack from all over the world.
> and i’m afraid about servers and bandwidth when more
> such apps will get more popularity.
Yes, that’s a problem that might come with popularity. But OTOH an offline browser like “knowledge” will take away requests from the live Wikipedia sites.
Library sources are far more likely to be accurate than the Internet, even though errors remain possible.
Websites shouldn’t need an API to extract information from them.
What makes a desktop application more of an applcation?
I find many web apps as useful or (increasingly) more useful.
Google, Yahoo, and Amazon all expose web APIs. Why not WikiMedia?
Library sources are far more likely to be accurate than the Internet, even though errors remain possible.
So, Wikipedia will be less accurate (as a % of content)or have fewer facts, overall, when compared to a paper-based encyclopedia?
Interesting claim. Want to back that up?
@Anonymous
Interesting claim. Want to back that up?
Books have already proved their reliability over the centuries, if anything needs “backing up” it’s your claim that Wikipedia is more accurate or that books are less accurate.
Wikipedia is great for those who understand it. OTOH, at this point the people I’ve discussed it with have generally previously either thought it was no different than traditional encyclopedias (bad) or have had a horrified kneejerk reaction to it (also bad). Integration with Google (via Answers.com) and Trillian (IM client) tends to increase the former.
People using KDE are probably the type who are more likely to be familiar with Wikipedia though…
Encyclopedias in general aren’t really suitable resources for academic research at the college level, btw. They can be good starting points and Wikipedia is quite useful for that as well.
A good academic library has professional librarians who’ve vetted the resources they make available (both print and electronic). Static web pages can be evaluated too, but Wikipedia articles are dynamic, meaning there’s a heavy onus on the reader to evaluate it–which is a fundamental skill they should possess anyway.
I see a new way of spamming if this initiative gets more wind in its sails. Say the new album from band Qwerty comes out, all the kids download it ilegally because their parents won’t let them buy it. They open up their media player, and the player downloads the Wikipedia info.
That Wikipedia info can be edited by anyone. By anyone. Also by Spammers. They put their spam/ad/links up on the Qwerty Wikipedia page, it gets loaded a few thousand times before the maintainers get their act together.
Bit farfetched maybe, but still, it’s not unthinkable.
What makes a desktop application more of an applcation?
A web app is used to fetch documents. It is not, at this stage, a fully interactive application, though a few sites are heading that way.
I should just be able to use networking and XML libraries to fetch and extract the information I want. All I should need to know is the format of the search query and results file.
So far it seems that this is not a huge problem. It will be interesting what happens if it *does* become a problem, but I guess they will use a SPAM filter similar to the one in email clients. Also, if they can be edited by anyone it also means that the SPAM can be removed by anyone.
Books have already proved their reliability over the centuries, if anything needs “backing up” it’s your claim that Wikipedia is more accurate or that books are less accurate.
Follow the thread. I did not make the original claim. I’m asking the ORIGINAL person to backup what THEY CLAIMED.
A web app is used to fetch documents. It is not, at this stage, a fully interactive application, though a few sites are heading that way.
Gmail or Google Maps aren’t fully interactive applications?
That said…except for location (one is local one is potentially remote)…I don’t see why APIs aren’t valid for web apps. Bits are bits…why ignore the orgin of some or special case them?
I should just be able to use networking and XML libraries to fetch and extract the information I want. All I should need to know is the format of the search query and results file.
That depends on the design of the APIs — something that the local or remote nature of the application doesn’t speak to.
Books have already proved their reliability over the centuries, […]
Like all the books about UFOs or Witchcraft…
Try getting it to run for more than 3 minutes without locking up. Fix the existing features before getting new ones.
Uh? I have amaroK running 24/7 without a problem. What backend are you using? Maybe GStreamer?
> I have amaroK running 24/7 without a problem.
Ack. No problems here either.
I was reading this at planet.kde.org
Nominees for Most Fancy User Request
Good evening. Sometimes users file funny, far beyond the capabilities of KDE, or crazy wishes. And I don’t talk about the “remove my neighbor country’s flag” evergreen. And the nominees for “The Most Fancy User Request” award are:
* Bug #34295 titled ‘No red spot on moon’ from a time where the RedHat QA department was still concerned about KDE.
* Wish #106880 titled ‘KDE music download shop like Apple iTunes’ for the best idea how to make KDE a successful business.
* The kde-devel thread titled ‘KDE development as a spectator sport!’ proposing to introduce bets on eg whether a developer will lose SVN write access.
Do you feel that one submission is missing? Then quickly nominate it before the jury pronounces its verdict.
Is sad to see how some developers are making fun of some KDE users that had done nothing but support them.
Amarok works nicely here, and is great. What hurts many people is that it is a KDE program, as K3B is 🙂
I find that Amarok copies the Contois Music interface but only iTunes got sued:
http://news.com.com/2300-1047_3-5756737-1.html
…except for location (one is local one is potentially remote)…I don’t see why APIs aren’t valid for web apps. Bits are bits…why ignore the orgin of some or special case them?
Why do some programs store numbers in text files rather than as binary?
What the vast majority of websites do is provide information. This should be done in a way that enables as many users as possible, be they human or machine, to access it as easily as possible. Devloping an API implies you have reached the limits of what this can achieve. It’s not that API’s aren’t ever valid, but displaying wikipedia pages in KDE apps does not require an API.
There’s no performance issue here. There is a platform transparency issue. Providing well designed XML means any platform that has an XML implementation can use it. Requiring the use of a different API for every website makes development that bit more complicated. Why use bits (which could mean anything) when you can use text?
I find that Amarok copies the Contois Music interface but only iTunes got sued:
Well, I know that if I decided to rob a bank, I’d probably choose one that had money rather than one with an empty vault.
Don’t you have to sue all offenders or none of them – in order to enforce your patent?. If they only sue iTunes then Apple can say you have to sue Amarok as well or they can claim that iTunes is based on Amarok’s design and since that design is not infriging then iTunes doesn’t infringe
I think this is an awesome way to make information more quickly accessible. Better yet! Incorporate Wikipedia entries into Beagle somehow for lightning fast searches. If the KDE team do more things like this, i might switch away from Gnome for a while.
As for the reliability/accuracy debate, Wikipedia is an awesome expiriment in a way, because it trusts that human nature will prevail in the disemination of information. Wikipedia is a great tool for gathering a working knowledge of something you were previously unaware. However, i wouldn’t deem it acceptable as a solitary source, but then again, nothing really should be.
> For your information, I use XFCE.
Ah, now *that* makes your ubiquitous anti-KDE trolling much better.
> Yeah, and I wonder where was the sense of humor when Nat
> Frieman was joking about developers dont taking showers
> and more than one of KDE devs.
I can only talk for myself (I wouldn’t call myself a full-fledged KDE developer, but still). I for one didn’t get offended. But I didn’t find that remark particularly funny or witty either. Nothing to talk about, really…
> I do not aprove this kind of comments becoming from
> anyone from anywhere, be it GNOME or KDE or etc.. etc…
Yes, of course, all you want is to make the world a better place. Given your troll track record I don’t believe you.
> And my advice to you is to no get blind by the fact
> that he is a KDE developer, that doesn’t make him
> untouchable
I agree with this, but I happen to disagree with you on the blog entry. And I question your motives.
My advice for you is to post
1) on-topic and
2) to leave the KDE threads alone.
It really gets old.
> > Ah, now *that* makes your ubiquitous anti-KDE trolling
> > much better.
> What trolling, I was just quoting what he said, nothing
> is invented, so is not trolling.
You may want to look up the definition of a troll in wikipedia. It’s not always about incorrect facts.
I don’t have to tell you that trolls often work with half-truths, or biased and distorted reproduction of facts. You already seem to be familiar with that concept.
I’m talking about your bad intention and your annoying habit to show up in many KDE thread and bad-mouth KDE. Probably to provoke a counter-reaction or to make KDE look bad.
Sometimes it’s the license troll (Qt is not free enough, Qt will never make it because it’s too expensive, bla bla). Now it’s a far-fetched story about the evil KDE developers (those users’ wishes *are* hilarious after all, and maybe even tongue-in-cheek! And besides it was one(!) KDE developer’s blog. You know what blogs are, don’t you? It’s not like it’s released as press announcement).
If you like XFCE or GNOME why don’t you find something positive to say about the respective developers in an XFCE or GNOME thread? Instead you think it’s your holy mission to let the world know how evil the KDE developers are. Your agenda is pretty clear.
This time you succeeded in provoking a counter-reaction. But I refuse argue with you any longer.
I should know better than to argue with a troll.
As far as I know, amaroK’s interface has _NOTHING_ to do with iTunes. Rhythmbox’s interface is itunes clone, yes, but amaroK has a whole different approach to UI than iTunes.
Assuming amaroK cloning iTunes without even looking at that might have these reasons:
1) They are both music players, so they must copy each other! This is how ugly patent claims work.
2) Assuming iTunes is the only way doing musical player right, so anyone would want to clone that.
both of which are false.
So, HOW EXACTLY amaroK clones Contois?
I’m interested, really.
Providing well designed XML means any platform that has an XML implementation can use it
That’s an API isn’t it?
The standard delivery mechanism of a website is HTML or XHTML as a response to a simple HTTP request.
If you want to access it through XML “commands” and get specified XML as a result, it is an XML RPC API.
I don’t understand how you can at the same time ask for an API and say an API is a bad thing
I’ve had this conversation many times regarding wilki and books…
According to ppl at Uni professors mark down work if wilki is even mentioned…(even when its not the only source)
Never got a valid explanation as to why…
Just that any darn soul can add information…. and that’s a bad thing….
What gets me is so can all these so called experts… add information (I am sure its more about wanting to feel elitist) n trying to show the world you have some black magic powers that others cant touch.
if information is writen in a book or electronicly and is the truth of the matter like a time line to what happened over the last few months… if the writen form gives all the pictures like for and agenst and others views and thoughts.
can someone please tell me why sharing as much information as possible is a bad thing?
Never got a valid explanation as to why…
They don’t get it yet. I expect that this will change in about 10 years. What is acceptable will depend on the specific source only, not the mode of presentation. People move slow.
can someone please tell me why sharing as much information as possible is a bad thing?
It isn’t…though it is one of those concepts that many people cling to…esp. the older generations.
I say this being an old guy. many years out of college.
No you can selectively enforce your patents.
XML or HTML, in this case, is just a representation of the state of some object.
The API spec is called HTTP and is really simple.
You reference an object through a globaly uniqe identifier called an URL, using the http: scheme. This object is then manipulated with the four methods
GET, for retrieving a representation of its state
PUT, for setting the state
POST, for sending data to be processed in other ways
DELETE, for deleting the object
by abusing HTTP and the POST method by just tunneling some XML based RPC you a) miss out on the excellent design of HTTP, and b) creates unneccesary overhead for your own API.