previous post: «Delicious statistics»
There is an enormous diversity in the landscape of social bookmark managers. Nonetheless most of the bloggers tend to stick to del.icio.us.
I rarely hear bloggers writing about another service than delicious. With this post I present two alternatives (RawSugar and Simpy) and want to prove that a spreading of users to alternative services could improve the world of tagging in general.
I recently tried out Simpy and RawSugar and thought about switching to one of them.
It is true that you are not really locked in to del.icio.us. You can export all of your bookmarks by just typing
curl --user username:password -o backup/delicous_backup.xml -O 'http://del.icio.us/api/posts/all'
But in a way you just feel like you’re locked in, isn’t it?
Why so?
Well, guess what? You are locked in! :-)
I think the point is that the advantages and disadvantages of the bookmark services aren’t known. I haven’t read much about it, have you?
I recently switched to RawSugar, after using del.icio.us for about 1 year. I have a little bit of experience using Simpy while using it for a little project. I will therefore confine myself to describe just this three services, if you have experiences with other services I would appreciate your comment.
Let’s start with the one you’re most probably using.
Advantages:
Disadvantages:
Joshua and his team are trying hard to limit del.icio.us to what it should do: Saving bookmarks and finding them:
“delicious is a tool, not a community. One reason i donut want it to be a community. some community behaviors are not good.” the “you suck” effect, and flame wars [Joshua Schachter at Carson Summit]
In fact del.icio.us has got the cleanest ui of all bookmark managers I’ve seen. On the other hand, I can’t imagine my dad using it. The whole terminology and the design are not the type of web page my dad would use. But then again this is an advantage. The fact that del.icio.us doesn’t want to please all users makes it simple for the ones it is made for.
The other disadvantages heavily depend on each other: When Joshua started his service he most probably took a few mysql tables and optimized his queries, when the demand started to grow he’s put in some cache here and there but then the big wave hit: in the last half year, the number of bookmark posts per day tripled. I guess he now is working on improving the performance rather than rolling out new features, as
“tags doesn’t map to sql at all. so use partial indexing.”[Joshua Schachter at Carson Summit]
The succeeding bookmarking service programmers already knew the problems of tagging applications before they started, Joshua and his team have to improve a running service. And then they switched to Yahoo’s location, that probably also helps slowing down the roll out of new features. The “private bookmark feature” is planned for about a year already, yet it is still to be released.
RawSugar is a bookmarking service run by a small company with a handful of programmers. It became beta in September of 2005.
Advantages
Disadvantages:
The fact that RawSugar started their service after the success of del.icio.us helped them to design their engine for big amount of data. They definitively have a good engine, As they never put public information about their engine, I’m not sure if I’m allowed to speak about details here, but they don’t just have a 3-table-mysql-schema :-) In fact I believe their engine would handle a wave of new users without stopping roll out of new features.
They have got hierarchic tags, and this is a very clever idea. First I wasn’t totally convinced about this feature but it has itself proved to be very helpful. I don’t like del.icio.us’ tag bundles. Tag bundles may be a good thing if you have 100 to 200 tags, but if you have 500 or 1000 tags (I currently have over 2000), it isn’t manageable any more. With RawSugar you can put a hierarchy in your tags while tagging an item. if you type “article>rant” you say that “rant” is a child tag to “article”. If you later tag a site with “rant”, the tag “article” will automatically be added.
A word about their terminology: At first it was difficult to map between the “delicious terms” and “rawsugar terms”. Sure, tags are still tags but somehow I was (too much) used to the whole terminology of delicious and took it as granted. On the other hand it’s good to see new viewpoints and that’s why I think new players in the tag market are a good thing. RawSugar aims to be not only a service for bloggers and geeks. Their UI looks more “traditional” like the one of Amazon, for instance: After you have searched your bookmarks after a certain tag you see a search field where you can type in another name of a tag and then hit “refine”. I think this is more “natural” than clicking the “+” signs at del.icio.us.
Finally they have a bunch of features:
Simpy is a one man project of Otis Gospodnetic. As he is/was a developer of Lucene he has a good knowledge of database systems and scaling. He has got a good engine too so I guess his service would still stand after a big wave of new users. Simpy seems to be online since early 2004 (according to the date of Otis’ earliest links).
Advantages:
Disadvantages:
What I particularly like about Simpy is it’s spirit: Otis is “one of us” giving away his spare time for this service and demanding nothing in return. In return, on Simpy’s developer mailing list he gets a lot of feedback from people that want to contribute, that is: write little applications, that use Simpy’s API. I am in contact with Otis: he is always very friendly and open for feedback. That said, he knows that his UI lacks some polish :-)
I’d say the killer-feature of Simpy is group tagging. While tagging you can click the checkbox of one of the groups you’re in. There’s an RSS-Feed of each group so this is a very useful feature. Ever tried to do that with del.icio.us? You have to log out and log in into the group account..
There are many many more bookmark manager services that I never actually used (I probably have signed up to about 50% to them just to get my favorite username in case they turn to get “teh r0×0r”), there’s a good list at listible.
In the end, there’s no real winner. And I like it that way. I suppose in long term each bookmark service will have it’s community.
The popular page on del.icio.us is ruled by entries about design and web2.0. I guess other services will attract other communities.
That said, I envisage two things:
Well, I highly encourage you to look beyond your own nose because there’s a big world beyond delicious. :-)
RSS feed for comments on this post. TrackBack URI
Thanks Philipp for telling me about those two other services. I’ll have a look at them later. Specifically I’m looking for a service that is properly translated, i.e. can switch it’s user interface to German or Spanish. You know of any such service?
Comment by Patrice — February 24, 2006 5:41 pm #comment-2284
Patrice: No, But Scuttle is an open source application that aims to be a delicious copy-cat. I couldn’t find out if it supports localization, so I’m afraid it doesn’t.
Then there’s Ning which helps you building a tag-based application, probably the service you can build with it can be in any language, but I don’t know.
Then, there are quite a bit of folks over at mildiez.net that speak spanish that could be capable to build such a service, they recently had a discussion about one of my articles.
Comment by Philipp Keller — February 24, 2006 5:59 pm #comment-2285
Sure, building such a service wouldn’t be difficult. It’s just that those narrow-minded Americans never think about I18N for their sites. It would be so easy…
Comment by Patrice — February 24, 2006 6:44 pm #comment-2286
Patrice:
I just found out that smarking.com has got an italian translation of the whole site.
I guess the big work lies in the translation of the page. A native speaker has to do it and he has to constantly add texts as new features flow in.
But nonetheless the multilingual feature could be very interesting: People could choose which language they want for the interface and which language they use for tagging their bookmarks.
I guess it also could automatically recognized which language the pages are in the people are bookmarking.
In the end the user could say that he only speaks his language and then he would see only pages in his language. This could help the bookmark services to get accepted by a broader audience.
Comment by Philipp Keller — February 25, 2006 5:59 pm #comment-2290
“It’s just that those narrow-minded Americans never think about I18N for their sites. It would be so easy…”
Patrice, I can’t speak for the other companies but RawSugar has 10 staff and only two are American/ The other eight are Israeli and our VP Engineering, who lives there now, actually was born and grew up in France before emigrating. So it isn’t American narrow-mindedness in our case.
I will dispute your assertion that I18N is easy. While the method for coding software this way is understood it is actually quite a bit of extra, difficult work; I can’t imagine Otis will ever be able implement it as long as Simpy is only him and it remains a closed source effort. When I was product manager at Sun for Java (iPlanet) application servers we had a whole separate team responsible for translating text and that’s somthing very small companies can rarely afford.
Phillip, thanks for such an informed, positive review of RawSugar!
Bill Lazar
http://www.rawsugar.com/links/billsaysthis
Comment by Bill Lazar — February 25, 2006 7:23 pm #comment-2291
Hi Bill
Thank you for responding to my comment. I wasn’t talking specifically about RawSuguar and not even specifically about tagging sites. It’s just that many sites never consider translation. And in my experience it’s often because of Americans who somehow often don’t realize the presence of people on that planet who do not speak English fluently enough.
It’s still my opinion that making the application translatable is not that big an effort – especially when planned from the beginning. Adding translation as an afterthought can be hard work, though.
The translation itself is a whole different topic. But look for example at how Google does it with their translation desk. That’s an innovative approach that spares them work and thus money but still gives them a very well translated site. Unfortunately this mainy seems to be used for google.com and not so much for other services (though Blogger is also translated quite well AFAIK).
And thus I conclude by saying that I disagree with your assertion that making a translation is difficult work. It may be a lot of work, yes. But not difficult. And when it’s done from the beginning it’s only very little work.
I’ve got (not too much but a little) experience with this BTW as here in Switzerland many applications are translated in more than just one of the three official languages.
Comment by Patrice — February 25, 2006 10:51 pm #comment-2292
@Philipp: Yes, translations would definitely help acceptance. I have a few friends who don’t read English and thus don’t read my blog anymore. And of course here in Peru knowledge of English is even lower.
And because I want to teach a course on how to use tools like del.icio.us, Wikipedia, etc. to improve teaching, I’m currently looking for a Spanish del.icio.us.
Comment by Patrice — February 25, 2006 10:54 pm #comment-2293
Philipp: nice overview. :)
Bill: never underestimate!
Patrice: I don’t know about narrow-minded Americans (Simpy was actually launched in Croatia, and had releases from China, Hong Kong, New York, San Francisco…), but I do know that Simpy has support for i18n. As a matter of fact, if you switch your browser to Croatian (HR), you will notice some portions of the UI translated to HRvatski (Croatian). So, in Simpy, it’s a matter of translation, not the matter of infrastructure and software to support it.
Comment by Otis Gospodnetic — February 28, 2006 4:57 am #comment-2304
Otis: Color me impressed!
Patrice: I agree that I18N requires less resources when implemented in the code base from the start. Unfortunately that’s rarely the case. We have not written off doing this but for now we have to devote our resources to more important (in our perspective) tasks.
Comment by Bill Lazar — February 28, 2006 9:11 pm #comment-2307
The attention.xml bit sounds interesting…how would you see this working? What would you consider to enough data within the attention data set? What data types would you forsee it containing to make it useful?
Alex.
Comment by Alex Barnett — March 3, 2006 7:24 am #comment-2313
Alex: I’m not really into attention.xml, but I guess attention.xml would have to contain at least some tag information. Somehow it must be extractable which tags I’m really interested in.
I can imagine the attention.xml not only coming from my feed reader but also from any bookmark service (delicious et. al.). Though I don’t know if that would be inside attention.xml spec.
Then the attention.xml would be parsed and probably a “top 20 tags” list would be built and compared with the “top tags” list of the different services.
The top matches would be presented to the user with an explanation why he should go to RawSugar or why he would best fit into Connotea.
Do you know Attention.xml? Would this thing be inside the specs?
Comment by Philipp Keller — March 3, 2006 8:11 am #comment-2314
scuttle is translated, have a look at the wiki.
Comment by igor — March 22, 2006 8:06 am #comment-2437
igor: good to know!
I found out that spurl is available in English, German, Spanish and Icelandic. Then on spurl you can even say that you just want to see bookmarks pointing to e.g. German and English sites (though I have no clue if that works and the site seems just to be down right now..)
Comment by Philipp Keller — March 22, 2006 8:33 am #comment-2438
I wrote a review of six bookmarking services, including del.icio.us, Simpy, RawSugar and BlinkList, at my blog. You can read it at http://jw.clawz.com/blog/2006/02/13/the-alternatives-ii-bookmarking/.
Comment by JW — April 17, 2006 6:57 pm #comment-2668
is there a world beyond here?
Comment by alex — July 27, 2006 8:28 pm #comment-5299