Anything and everything that interests me might show up here.
My interests include Jazz, (Auto)Road Racing, NetMeeting, E-Learning, Zope/Plone, Creative Problem Solving and lots of other stuff.

Last update: 15/05/2008; 5:30:29 PM.

May 2008
Sun Mon Tue Wed Thu Fri Sat
        1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30 31
Apr   Jun

Home

    Subscribe to "Brian Sullivan's Random Musings" in Radio UserLand. Click to see the XML version of this web page. Click here to send an email to the editor of this weblog.


    Recent Posts

     5/15/08
     5/14/08
     5/9/08
     5/9/08
     4/28/08
     4/23/08
     4/21/08
     4/14/08
     3/5/08
     2/25/08
     2/13/08
     1/31/08
     1/28/08
     1/17/08
     1/12/08
     1/7/08
     12/29/07
     12/27/07
     12/23/07
     12/19/07
     12/15/07
     12/15/07
     12/13/07
     12/12/07
     12/12/07
     12/11/07

    Lijit Search

    May 15, 2008
     

    Regular Twitter outage

    Twitter down again!
    5:30:27 PM        comment []   
    May 14, 2008
     

    Twitter forked again

    So it seems the recent departure of top technical people has not helped Twitter's stability (maybe all that money will?). The site is totally unavailable at this time (645 edt) and has been for at least 1/2 hour. Better than sort of working I guess?

    Strangely enough the server that is www.twitter.com is pingable (I would have thought that would not be possible at any time --makes them very DOS vulerable).

    6:47:11 PM        comment []   
    May 9, 2008
     

    Google maps test

    I have been testing -- trying to find a way to put Google maps on my RU weblog. This seems a problem for all weblog type websites (in fact any content management system based websites) though not just RU.

    Having one map per page like here is possible in a contrived sort of way but I would like to have a mechanism to add a map where the map is totally described and created within the blog entry.

    The way maps are currently created they require generation within the head portion of the page (requiring foreknowledge of the parameters and names of maps on the page). That doesn't seem to fit with the weblog model which presumably should allow for multiple maps within a weblog entry and 0 or more weblog entries containing maps on a page. Furthermore the numbers of maps is dynamic and variable.

    Queries in the Google Maps support forum have yielded no results. In fact there is no indication even that the problem is understood.

    The only potential solution I can think of is to force the tags describing the map locations to contain a common class and some description of the setup parameters required for the map, and then search for all class members in the head Javascript, creating the appropriate maps. I haven't tried that yet and I am not sure my Javascripting skills are up to it.

    Surely someone out there has run into this problem before and either determined a solution or decided that there is no sensible solution?

     

    Update: I am investigating two possible avenues -- one is to use the "Static Map API" which gets you a Google map jpg, gif or png -- maybe I could hyperlink that to the equivalent interactive mape?

    Another is to put the map on a separate html page as it seems to want and use in Iframe tag in the weblog entry.

    10:35:58 AM        comment []   

    Twitter $$$ continued

    ReadWriteWeb has an article discussing possible monetization models for Twitter. These are based on current traditional strategies being used (or attempted) at other web based service companies.

    From my reading none of these will work given the current API model of Twitter which seems to supply the complete means to replace the UI. I use Betwittered (an iGoogle gadget) that does just that and could be headed to have a better web UI than Twitter has. Effectively others can build circumventing UIs hijacking any traditionally based revenue model. The only way that Twitter can retain control is to shut down or limit the API.

    Any other ideas for Twitter revenue?

    10:19:30 AM        comment []   
    April 28, 2008
     

    End game for Skype - $0 revenue? What about Twitter?

    Evan Williams poses an interesting question on Twitter:

    "Does Skype make less money the more users they get? (Because previous users no longer have to be Skype-Out minutes to talk to them.)"

    It seems obvious that there is an issue here. Assuming the current business model (selling in and out minutes for calling non Skype users with an otherwise free service) and ultimate success (no non-Skype telephone users), Skype's revenue would be $0. That hardly seems like a good business plan or goal.

    As Evan's question indicates every new Skype user implies a potential loss in revenue from existing users (they will no longer have to buy in our out minutes to converse with the new user). It also has a potential gain in revenue as the new user buys in and out minutes to converse with non Skype users in his/her circle of contacts.  At some point the loss in revenue will in fact exceed the gain. I am not sure where that would be. Somebody with vastly superior mathematics skills and business/economics knowledge can probably comment.

    Mitigating factors would seem to be:

    • Skype could continue to increase minute in/out costs to equalize the revenue (but not to infinity)
    • Skype could find another balancing revenue plan (it could supply the world's telephone needs and make money some other way)

    It would seem that the current revenue model is a kind of reverse pyramid scheme. There may well be some sort standard explanation for this apparent paradox I am not aware of but Evan's seemingly off-hand twitter raises a whole line of interesting questions.

    One of the most interesting questions for me is why is Evan (currently the man behind Twitter) raising/pondering this issue at this time. Is this somehow a clue regarding the conundrum of the still not determined Twitter business model?

    11:55:59 AM        comment []   
    April 23, 2008
     

    Twitter is saved!

    Arrington reports that amateur hour is over at Twitter- that Blaine Cook who apparently was responsible for scaling at Twitter has left. Somehow he equates that with solving all of Twitter's problems. He is much closer to the situation than I and he may be correct.

    But somehow I doubt it. From the number of problems, the way they have been handled I am guessing that the source is much more pervasive than just the lack of ability of one person. I think the general incompetence likely runs up and down the organization, and/or the base technology is just not up to snuff.

    Add that to the fact that there still seems to be no way to make money from Twitter or any sign that there is a revenue plan for the future (notwithstanding the Japanese ad trial), I am guessing that tension  in the organization is very high. I think the window of opportunity is shrinking and everybody there knows it.

    11:54:09 AM        comment []   
    April 21, 2008
     

    Tweetless

    "What we have here is failure to communicate."

    I have talked before about my feeling that there seems to be something rotten at Twitter at the technical level. The latest problems on the weekend seem to confirm this.

    This current situation though is much worse -- most users don't see any indication of a problem. Messages are just being lost or not displayed. Not only that but Twitter seems to be failing at the corporate/PR level as well. No communication with users, no messages explaining what is going on, just a deathly silence.

    Where is Twitter's management and technical staff? All playing in the weekend California sun, counting their future fortunes?

    10:34:56 AM        comment []   
    April 14, 2008
     

    Robert Scoble and Twitter

    I started to do the mental arithmetic on Robert Scoble's claim that he "follows" 20,000 people in Twitter. Pondering how many messages a day 20,000 can generate. If I assume 1 message per day per person (from the people that I follow that would be way low) that would be 20,000 messages per day to read, scan or somehow process. Since there are 1440 minutes in a day that would imply 13.8888889 messages a minute -- one every 4 to 5 seconds.

    Given that there must be some time in a day for sleeping, eating and other necessities and given that Robert has many other things to do other than pay attention to Twitter the message process rate must be way higher (perhaps two or three or more times as high). So that is approaching  1 or more messages a second. I know there are claims that Robert is a cyborg (though from personal knowledge that seems not the case) so that seems not a plausible explanation.

    I think then that Robert's claim is somewhat suspect (of course his definition of "follow" may be different from mine -- or my math may be wrong).

    So come on Robert 'fess up. Are you really "following" 20,000 on Twitter - at least in any real sense of the word "follow"?

    10:39:49 AM        comment []   


    Click here to visit the Radio UserLand website. © Copyright 2008 Brian Sullivan.