 |
|
| FriendFeed is the easiest way to share online. Learn more » |
Roderic Page › Likes
Guardian gives Google Sidewiki 6 mths until internet roadkill. http://www.guardian.co.uk/technol... I am neg as it is part of G toolbar which I hate
Lars Juhl Jensen, 'Mummi' Thorisson, Roderic Page and 2 other people liked this
Interesting that the focus here seems to be on webmasters losing control over how their websites are seen. Errr hasn't that already happened. Greasemonkey? Stylish? Just in those cases the user has the choice about what is put in place - reducing the potential for spam
- Cameron Neylon
I'm waiting for that Google wave thingy, Cameron - sounds much better. For people like me for whom words like "greasemonkey" are a mirage, we want some kind of application for us that will do it....I suppose. BTW is Twitter full of spam today, looking at the trending topic list? (will not repeat the words here just in case they all follow).
- Maxine
I like the idea of a global, shared annotation system. I have lots of notes in, say, Evernote, about data records that are wrong, or could benefit from some extra links. But these are not shared. If I moved these to Sidewiki then chances are others might use them, including (gasp) the person responsible for the original page. Indeed, web sites could subscribe to ATOM feeds for individual pages to get feedback and annotations.
- Roderic Page
Well I wouldn't mind that at all (on a much less technical level) but would Sidewiki be the way to do it? Not according to the Grauniad and others (eg John Batelle had a neg post about Gsidewiki too - concluding that Yahoo! should do it instead. ?!). For me, I'm not installing that Google toolbar again, whatever! I like some google apps but not that one. (or any toolbar along same lines by other companies.)
- Maxine
Inappropriate name or not - this stuff has serious potential!! As Rod alludes, like the other Google Data APIs this based on AtomPub (http://www.atomenabled.org). I'm not terribly keen on the toolbar / separate install thing, but the principle is good and presumably one could write JS widgets to make this work just like regular comment threads (like Disqus: http://disqus.com).
- 'Mummi' Thorisson
Also, would be kinda nice if there was a way to leverage the same/similar infrastructure to gather and archive annotations on bio-db entries to one or several repositories (as to not rely on Google for storage and, ehm, not do Evil). If this could be done + without the toolbar, I could agree to Todd's sentiment ( http://ff.im/8CnSo) when he asks if this might be "the answer for scientific databases wishing to add community annotation features?".
- 'Mummi' Thorisson
@Maxine - if you could leave a SideWiki comment on a site without having to use the toolbar thingy, would you be more positive, and also would it matter to you whether the comment was hosted on the site or with our overlords @Google?
- 'Mummi' Thorisson
The idea itself isn't bad. The implementation on the other hand, means it won't catch on.
- Deepak Singh
Google is not the first company to do website annotation like this. All the others were killed by spam and, although Google must have some super-strength anti-spam algorithms, I predict that this will be too. For now though, sidewiki is going to be built into Chrome, which slightly alleviates the toolbar issue. Also you can already view sidewiki without the toolbar. Just make a...
more...
- Matt Leifer
@Deepak - the UI implementation you mean, or do you have reservations about other things?
- 'Mummi' Thorisson
Thanks, Mummi - yes, I'd have a go at using it if I didn't have to install a toolbar first. To me as a user, what I like is the ability to integrate all the conversation in one place, does not matter to me where. But as Deepak says, the devil is in the implementation....Spam also a major issue - an increasingly major one it seems to me, as spammers increasingly learn the increasingly sophisticated anti-spam algorithms ;-(
- Maxine
Mummi, the interaction model is a show stopper for a lot of people, and integration into chrome is not going to help (since such a small number of people use it). Personally, I believe the ability to annotate and collection information on a web page at scale should be part of some core web standards, otherwise it ends up becoming interesting to just a few people. And yes, spam is going to be huge issue (regardless of model)
- Deepak Singh
@Deepak - regarding spam: surely it will help that you have to at least have a Google account to use this.
- 'Mummi' Thorisson
Mummi, perhaps. But that'll get circumvented some day. That's not my main problem. First they need to get non-spammers to use this
- Deepak Singh
Still, given that Google are in fact building this on standards (more or less), that seems like a step in the right direction to me. Like Todd, I've been for some time interested in something like SideWiki for annotations of individual database entries. Not merely annotations gathered by each individual db (many resources will not have the resources to build this) but more likely one or several specialized bio-annotation servers.
- 'Mummi' Thorisson
..If we have Google SideWiki + a number of these bio-annotation servers implemented as AtomPub servers, theoretically one could aggregate feeds w/ comments from whichever sources would be appropriate. If SideWiki is full of rubbish, don't aggregate it, and stick to the two bio-annotation servers X and Y who are used by the specialist 'non-spammers'.
- 'Mummi' Thorisson
Mummi, one could argue Anyone should be able to collect annotations across the web and pull in the parts they need to. So in that context Sidewiki just becomes another interesting part of the equation. The fun part is the core sources, and those sources opening themselves up to annotation, which to me is the more important problem. Is making sure that source data has APIs, allows third party annotations and or curation, etc
- Deepak Singh
BTW it seems Google are building on some of the ideas explored in the W3C Annotea project: http://www.annotea.org / http://www.w3.org/2001...
- 'Mummi' Thorisson
One blogger's opinion: http://thefowle.livejournal.com/391464... "Annotea was never tasked with designing syndication-- their assigned role was to define what document markups we may need. SideWiki appears aimed almost exclusively at syndication, at sharing the most primitive of content. The hybridization of these two technologies is the sweet spot hyper-geeks dream of, stirred in amidsts all the diverse medias of the web-- web pages, sound, & video."
- 'Mummi' Thorisson
@Deepak - agreed, for the most part: decoupling the annotations from the sources seems key to me here. Many data sources will, yes, have APIs for retrieving data, but no need for or interest in opening up (harvesting?) for external annotations themselves. So, enabling annotations to be 'stuck onto' *any* database entry identified by a URI using 3rd party (SideWiki, whichever) is crucial.
- 'Mummi' Thorisson
there I agree. You can address this from two directions. As I said, the SideWiki idea is not bad in itself. It's the interaction mode which is going to be the challenge (of course I remember saying that for the original Google Reader)
- Deepak Singh
from IM
Well if they develop it like GR, that would be good. However, they do add features to GR which only work for Blogger, eg the "following" function - unnecessary if you are already signed up to GR. And none of these blog platforms seem to integrate with each other (blogger, Wordpress,Typepad) although they all integrate with Twitter, FB and sometimes FF. Has anyone noticed the freq. crashes if you try to make a comment to a Blogger blog using your OpenID as opposed to your Google ID? All a bit irritating.
- Maxine
I disagree http://twitpic.com/jyhjx RT @gregtyrelle #Excel is the universal data analysis tool in science
 |
Michael Kuhn, Christopher Fields, Greg Tyrelle and 15 other people liked this
that picture has made my Friday
- Simon Cockell
I have seen this in real life, and when I offered to write a script to do it faster and quicker, that you could re-use again, the response was, "but why?
- Frank
which also reminds me of someone who was checking 7 different websites every week getting upto date information a certain identifier. When asked, if I wrote you a workflow to do the checking automatically and only alert you when the information you are looking for has changed, do you think that would be useful?, the response was "No, I do not".
- Frank
LOL - fantastic!
- Daniel Swan
Best picture I've seen all week.
- Chris Miller
ROFL!
- Yann Abraham
I stand corrected :) I haven been sent a word file with sequence data for a while. Must be the realization that pdf files are better for that kind of thing...
- Greg Tyrelle
Test: Annotation of http://www.ncbi.nlm.nih.gov/pubmed... with Google SideWiki. - http://lindenb.tumblr.com/post...
 |
joergkurtwegner, Steve Koch, Roderic Page and Neil Saunders liked this
Very cool, but need exactly the same URL to see the annotation. http://www.ncbi.nlm.nih.gov/pubmed... doesn't show the annotations. Given that there are a number of URLs for the same PubMed article (never mind other identifiers such as DOIs) it's going to be important to use the same URL.
- Roderic Page
Roderic, It's at least better than Annozila ( http://annozilla.mozdev.org ) . Because the later don't use the GET query , so http://www.ncbi.nlm.nih.gov/pubmed... was always http://www.ncbi.nlm.nih.gov/pubmed . Moreover, there is an API, so we can imagine a tool updating the information from connotea/citeulike/delicious/etc...
- Pierre Lindenbaum
Oh don't get me wrong, I think it's cool. I'm just thinking out loud. In order to retrieve comments via the API (i.e., using the ATOM feed) I need to get exactly the same URI as the one you used. If I land on the page via http://www.ncbi.nlm.nih.gov/pubmed... I won't see your annotations. So, it assumes people will consistently use the same URIs
- Roderic Page
No problem Roderic, that's what I understood :-)
- Pierre Lindenbaum
You can partially solve the issue by having web page producers insert < link rel = " canonical " href = " http://www.ncbi.nlm.nih.gov/pubmed... " / > or similar in the header section.
- Peter Ansell
top google hit for the top 1000 gene wiki pages: wikipedia 391, genecards 340, ghr.nlm.nih.gov 23, ncbi.nlm.nih.gov 13... (re: http://friendfeed.com/rdmpage...)
joergkurtwegner, Nakao M., Pierre Lindenbaum and 4 other people liked this
Final score (out of 9694 pages): genecards 4126, wikipedia 3940, ncbi.nlm.nih.gov 150, yeastgenome.org 82. Hat tip to the genecards folks...
- Andrew Su
http://stackoverflow.com/ is truly excellent idea, community is v.helpful, even more than on IRC.
Roderic Page, Daniel J. Pritchett, Tom Morris and 4 other people liked this
I'm telling you, this is highlight of the week at least, all those badges, tags, comments, reputation! Imagine the same system on FF :-)
- Piotr Byzia
from twhirl
I really like stackoverflow, but not sure you want a reputation system on FF. I kinda like the informal aspect of the whole thing.
- Deepak Singh
Hmm, you're probably right, I'm just keen to see any system reducing information noise, 'hide' is not enough IMHO.
- Piotr Byzia
from twhirl
@Piotr You mean something like this? http://tinyurl.com/89a5qu Also, I think Stack Overflow is awesome, though I'm a bit perplexed at why it works. http://tinyurl.com/c7n9bc
- Chris Lasher
The other difference is that on FF, your karma is contextual to an individual. In other words you control whom you hide, etc. The stack overflow system is global.
- Deepak Singh
@Chris Yes, I'm pretty sure tagging is a high priority on the FF developer's TODO and it will be convenient ones it's here. What I like about SO are suggestions (when writing your question) and community editing (tags, comments, questions, answers, pretty everything) so that there is endless improvements in content.
- Piotr Byzia
@Deepak Karma is no harm to anybody, ideally, you would be able to filter content by a particular karma level (i.e. 'show me comments from people with karma >1.5k) Add to this tags (... with tags: R;motifs) and create an RSS feed based on that...
- Piotr Byzia
@Chris Great post about SO! I agree on all points you made there, especially on "psychological carrot" in getting karma. But I would say that programmers active on SO are just aficionados of their tools (AKA geeks ;) ). Quite similar to who you can find on FF in some of their rooms ;-)
- Piotr Byzia
Dont like reputation based groups. Instead I prefer the old fashioned email based newsgroup. When I see a two page reply that is the perfect solution ..thats sheer altruism with no strings attached. I still turn to comp.lang.python or some googlegroup before SO
- Hari
I can't use mail based groups any more, even small internal groups. Too insular, and too much noise since email is such a bad medium here. Plus there is very limited google juice. The day of going to a site are going way.
- Deepak Singh
Wow, Hari, you dig deep in archive, this thread is from March ;-) I follow c.l.p. but only through RSS, for getting quick answers SO is the best, and quality is good enough to get things done.
- Piotr Byzia
I agree with Deepak, if mailing lists -- only digest version.
- Piotr Byzia
Nature Precedings and the 3rd International Biocuration Conferance - http://precedings.nature.com/collect...
Abhishek Tiwari, Hilary, Cameron Neylon and Roderic Page liked this
Slide from talks and posters from the 3rd International Biocuration Conference (April 16-19 Berlin, Germany) are being made available on the Nature Precedings site:
- Frank
OSCON 2005 Keynote - Identity 2.0 - http://identity20.com/media...
Roderic Page liked this
One of the great presentations. The first time I saw this I was blown away by the style (ideas aren't to bad either...)
- Roderic Page
Other ways to read this feed:
