> > Yes, I can see a lot of use for having Google on the desktop.
> > But how can web service-based Google application developers
> > compete with HTML scrapping apps that doesn't have to pay
> > Google any money?
>
> I couldn't really parse the question, but have you tried to
> build anything interesting out of screenscraping? It sucks,
> free results or otherwise. Paying some bucks for a regular
> grammar is worth it.
Yes, I have and screenscraping does indeed suck in general. HTML
returned by Google is, however, fairly consistent and highly scrappable.
> > b. offer web services-based features not available through HTML.
> >
> > I don't see (a) being very effective. (b) will work but
> > requires fairly significant investment.
>
> How so, significant?
Implementing new features require new data, new data structure, and/or
new code. The 'significant' part is just my eyeballing size of task
involved. If such features are going to be available only through web
services, I suspect the decision to commit the resources to implement
those features will have to rely heavily on faith and hope.
> > 3. Google appliance
> >
> > If this is the only way Google can justify their API, wake me
> > up when there is news.
>
> Why the scorn? have you concluded that a webserviced Google
> has no valid proposition? Tell us more, please.
Bill, I am not scorning Google nor web services. I just want to see if
there are real and substantial business cases for web services using
concrete examples like Google. So far, I have not heard anything that
justifies the amount of hoopla over Google's SOAP API.
Best,
Don Park
