This is a purely technical blog concerning topics such as Python, Ruby, Dart, Linux, open source software, the Web, and lesser-known programming languages. Ad maiorem Dei gloriam inque hominum salutem.
Monday, April 07, 2008
Python: Google App Engine
I just found out about Google App Engine. Does anyone else feel as overwhelmed as I do? This changes everything for us Python Web guys.
I don't see what's overwhelming. They've put out Yet Another web system with some odd constraints. Unless you want to use another database over the web, your app will be locked into a proprietary database system that you can't run (or even test with) anywhere else.
For GAE to change the game, I would think they'd have to make it a lot easier for people to create web apps (not "here's some of django and some other CGI you can use").
GQL is especially weird, since it's a limited SQL-ish thing that no one else supports or necessarily will support. I'm a fan of the new SPARQL language for querying RDF, partly because SPARQL is the kind of standard system that google *could* map to their database. RDF predicates (aka google datastore 'keys' aka SQL 'columns') are more precise than any of the other systems, so it's never lossy to adapt an existing DB system to RDF.
Doesn't facebook have an FBQL that's also an arbitrarily reduced version of SQL? Maybe it's not for the same purpose at all, but just the presence of these things is training everyone to accept a new, proprietary query language for every system we deal with :(
I do think it is a big deal. Having Google throw it's massive infrastructure and support around creating Python web applications is going to change everything...IMHO.
5 comments:
Indeed.
This could make Python the next PHP. (And maybe in the bad ways too ;)
Open source web applications could, like, really work. Normal people could install a web application like they install normal applications.
I don't see what's overwhelming. They've put out Yet Another web system with some odd constraints. Unless you want to use another database over the web, your app will be locked into a proprietary database system that you can't run (or even test with) anywhere else.
For GAE to change the game, I would think they'd have to make it a lot easier for people to create web apps (not "here's some of django and some other CGI you can use").
GQL is especially weird, since it's a limited SQL-ish thing that no one else supports or necessarily will support. I'm a fan of the new SPARQL language for querying RDF, partly because SPARQL is the kind of standard system that google *could* map to their database. RDF predicates (aka google datastore 'keys' aka SQL 'columns') are more precise than any of the other systems, so it's never lossy to adapt an existing DB system to RDF.
Doesn't facebook have an FBQL that's also an arbitrarily reduced version of SQL? Maybe it's not for the same purpose at all, but just the presence of these things is training everyone to accept a new, proprietary query language for every system we deal with :(
Drew, very, very well put.
I do think it is a big deal. Having Google throw it's massive infrastructure and support around creating Python web applications is going to change everything...IMHO.
Have to agree. Google don't do things by half. I just hope the Emacs lads integrate it properly :-;
Post a Comment