CommenTag | official blog

I have been working hard lately to get the new server running. That’s exaclty the moment when I really appreciate coffee.

I have to say that I’m a proud owner of a Senseo machine, and I enjoy it! It may not be the best coffee in the world but the machine is simply brilliant. Push a button, put a coffee pad, push another button and that’s it! It soooo easy to make a cup of coffee. My machine also has the “Auto-off” feature (apparently not included in the first versions) which shuts off the machine when not used.

Those Senseo guys just follow the KISS (keep it simple, stupid) principle and got it all right. I’m pretty sure that if they didn’t invented the Senseo, Apple would have. Yeah right, Senseo is to coffee machine what Mac is to computers. That’s how everything should be in life: just simple and easy to use.

Oh, yeah I almost forgot to tell you about the server. Apache, mySQL and java (no silly joke with coffee) where configured and data was migrated. www.commentag.com is now running on the new server. Don’t hesitate to contact me if you notice any problem.

Next step: add the last features, nlp integration and bugs fixes in API before open beta

Here are the main lines of the last week activity concerning NLP :

- several attempts to improve performances of the research of a specific tag’s semantic neighbours : still in progress

- testing the execution on a local server : done

- adding new fields inside database for learning : added tables for tag groups, tag and tag-tag frequencies + first design for an implementation of incremental learning using co-location an co-occurences measures on every tag

- testing the use of OpenCyc as an ontology (www.opencyc.org) : can be added to the neighbour words intension level

That’s all for now… See you next week !

Yesterday night (Thursday 13th of March) I had a one hour skype call with Scott Rafer, CEO of MyBlogLog and founder of Lookery.
We have definitely a lot to learn from Scott’s experience. And as expected it was a very useful conversation.
Scott has been very kind an gave me a lot of advises that I want to share with you below.
If you are also in the process of bootstrapping your company (hi guys from @minibar!), they might be more than interesting.

He wouldn’t invest though as he is already in too many businesses.
However he is more than keen to help us and that’s priceless.

Sum up of his feedback:

• He loves the idea
• He loves the way we implemented it so far (letting users to create on-the-fly their own taxonomy by selecting multiple tags)
• We are not ambitious enough
• Although we can start with tagging comments, we should extend our system to any kind of small content (”bunch of microdocuments”). We could tag twitter messages, polls of polldaddy.com, presentations on slideshares.net, … plenty of such little things.(personal note: the web so far has been all about interconnecting meaningless html pages. The future of the web would be an interconnection of smaller entities (an address, a contact, an event, a bank statement, a poll, …) but which convey a defined meaning. The thought of Tim Berners-Lee, inventor of the WWW, is worth reading on that topic).
• We should change the name “commentag” to something else otherwise it’s gonna restrict us to deal with other things than comments.
  • His proposals:
    • tagtronica.com “it’s 4 syllables and 3’s normally but limit, but it’s not terrible”
    • tagtize.com “is shorter but it bugs me”
  • My proposals:
    • centraltags.com (I really like this one)
      
      Quick draft for a logo
      
    • tagtoes.com (funny enough, my colleague David told me this afternoon that a toetag is a tag we used to put on the toe of the dead people (with their names and other infos). That might be a good name then as we will tag content that would be lost in a mass of other content, and if you are lost, then you dead somehow).
    • tagtoo.com (unfortunately the domain is already taken…)
  • Any other proposal? Feel free to let us your thoughts below!
• We should not care at this stage about any business model
• Our only concern now should be:
  • raise £300k to put 5 guys (included ourselves) full time on the project
  • Only address the mass market (don’t care about B2B)
  • Once you get into people referrer logs, you’re done. (ie. for the moment, we do not exist yet)
• When asking about moving to SF he answered: “San Francisco (SF) is a great place for funding, but terrible place for hiring”
• What a lot of companies do is setting their commercial management in SF for the networking and let their CTO and engineers in another country.
• We should consider using a virtual hosting platform like the one proposed by Amazon. @Arnaud: please go check it out!
• Next steps:
  • Define a roadmap for the different phases of our project
  • 1-2 page business summary. Not about the features, but more about the market size, the opportunities, … (ie. SWOT analysis)

After thoughts:

While seeking for a name, I land on tagfs.
A name a bit too techy but the idea is all in there: We could imagine a file system for the web which would be mainly based on tags.
tagfs would be a platform containing only i-nodes (ie. only reference to be able to find the content back - like in the unix filesystem).
Any web service would be able to register any atomic entity (a poll, a presentation, a picture, a bank statement, a booking, a thought, …) in that platform. Allowing users to gather at a glance all their stuff concerning, for example, their last holidays (pictures, contacts, bills, reviews of hotels, …).

To be continued…
(stay tuned)