500.000 ebooks gratis da Google per Sony Reader

L’annuncio è importante: Google, grazie al progetto Google Book Search, ha messo a disposizione gratuitamente tutta la sua collezione di libri di pubblico dominio, ma solo per Sony Reader.

Per accedere a questa enorme libreria, bisogna scaricare il software apposito dal Sony eBook Store (che, però, si può installare in Italia solo con un trucco, vedi per esempio post).

Personalmente ho provato scaricando Flatland sul mio Sony Reader, e devo dire che il risultato è buono (anche se il libro è stato digitalizzato, quindi con tutti gli sporchi e i difetti del caso), e l’idea di avere a disposizione tale enorme numero di libri alla distanza di un click ha dell’incredibile.

La notizia è ormai diffusa e potete trovare i dettagli su tutti i maggiori siti di informazione; è interessante osservare come questo annuncio venga messo in diretta concorrenza con Amazon e il suo Kindle, e questa concorrenza non potrà, secondo me, che fare del bene a tutti noi amanti del libro nella sua forma digitale.


Recently I noted that I do not have a lot of comments on my blog (around 60 on 120 posts from 2003) but anyway I would like to some more feedback, so I installed the very-well-done plugin WP-PostRatings: now even my lazier reader can rate any post or page with only one click! 😉

Remember: in general, who writes does it because likes it (and likes also having some form of reward), but he/she cannot do better without your help so, if you like some post in some blog, please invest some minutes and do comment it!

The analysis phase: time to grow up?

When small software companies get bigger they embark on what can be a bumpy ride of change. One of those changes will probably be to do with the way they tackle the analysis phase of the software development life-cycle (SDL). Just to be clear, when I say “analysis phase”, I mean the part before coding starts i.e. requirements elicitation, analysis and system specification.

Typically (although I am sure that there are plenty of shining examples where this is not the case) small software companies with a handful of developers, where the entire SDL for a project is covered by one or two developers, tend not to have a formalised analysis phase. Why is that?

Continue reading The analysis phase: time to grow up?

Welcome to James Kerr

I welcome James Kerr, a friend of mine and a colleague, on my blog: he’ll write about requirements engineering and systems analysis, i.e. about everything that should be done before starting to write code (in the agile world, too, even if with less rigidity).

It is my hope that more of my friends will follow James and together we can create a place where people can discuss software development problems, and find some good solutions.

Visual Studio.NET 2005 Web Test problems

A collegue asked me to solve some “strange problems” with VS Web Test: hidden fields are treated in a semi-random fashion and not all http request seems to be tracked effectively. After some googling, I’ve found the following links:

Web Test Recording

Automatic Hidden Field Tracking

Web Test Authoring and Debugging Techniques (the paragraphs Handling View State and Other Dynamic Parameters and Fixing __EVENTTARGET and other hidden form fields modified by JavaScript)

Can hidden field detection be disabled in recorder?

Summarizing, VS Web Test doesn’t behave like a proxy to intercept request, but instead it hooks inside Internet Explorer’ event system and so tracks only calls corresponding to user-generated events: that has the advantage to skip all related requests (mainly static files, like js and css,) but has the side effect to skip the Ajax calls not directly related to user actions.

So here comes a simple question: in a world that is going more and more toward an application-in-a-browser mode (so essentially based on Ajax), why Microsoft opted for this solution that in not Ajax-friendly? The other side effect, i.e. the fact that it works only with Internet Explorer, is less misterious to me.

The second problem, related to hidden fields, is quickly explained: it’s a bug. It originates from a smart feature: extraction rules, that helps the tester cope with hidden fields containing session-related data (like view state): the bug causes all hidden fields to be treated like session-related ones, causing in practice that their values are lost, so recorded tests cannot be run without manually editing it.

There are two workarounds to these problems:

  • you can use Fiddler, a proxy-based recorder that can save recorded sessions in VS Web Test format; a smart tool with a strange user interface
  • if you want to record with VS Web Test, you can use a trick: record normally, the save the recorded session in Visual Studio before stopping it in Internet Explorer, then stop it, and close the file responding “No” to the save request.

In both case, the workaround can be not functioning, because sometimes Fiddler, which works like a charm, cannot save the session in VS Web Test format, giving an error not clear related to XML correctness; on the other side, a session successfully recorded with VS Web Test doesn’t work with the web app without editing its source form.

More to come, if I find some clever solution.