JRoller

I have no idea how many subpoenas Google gets per day but my guess is that they get a fair bit of them. I also don't know how they respond to them but I'd wager that they comply with vast majority of them. What I'd like to know is how Google deals with out-of-US subpoenas. Do they comply?

I think you could argue that, due to Google File System's (GFS) distributed nature, you could request a subpoena from a local judge in any country in which Google has a data center because Google would have to prove that a chunk of data belonging to someone is not hosted in that country. That's very hard ot disprove since I doubt that Google keeps historical track of their chunks. The way GFS works, a master keeps tracks of chunks but it doesn't keep a history of where each chunk was. So a clever lawyer could try to subpoena anyone's account given that Google has a data center in their country.

For example, Google has a data center in UK. Could US person's account, be subjected to a UK issued subpoena if a lawyer uses the logic I described above? Maybe.. there's no way to prove that a chunk of your data is not hosted in UK.

Distributed systems ("cloud") are a privacy nightmare waiting to happen, and I think that EU privacy commission will tackle this issue at some point. "Cloud" computing seems to be illegal (to a certain degree, in strong-privacy countries like Germany).

More info here: http://www.jroller.com/MasterMark/entry/google_teh_evil_cloud_economics

Scala:

I seem to remember some "Is it ready for the enterprise, programming in the large, giant consulting firms with giant RFP's yet/soon?" threads/blogs which i can't find. Also some "average developers get python, ruby and java just fine, but they're not going to get scala and clojure".

I've heard chatter from friends and on the tubes that they're going to convert their humongous java dev teams "soon" but the main knocks are:

- problems in plugins for eclipse, intelliJ, and netBeans, (assuming it survives in some supported form

- it's taken so long to get 2.8 beta out: Well, nobody said it would be easy.

- OO type inference, FP, scripting, what is scala?; you can use scala as a scripting language, use a small language subset, apply regexes and maps and cut up apache logfiles easily just as in python or ruby.

- Or you can spend a lot of time reading about actors(akka, kilim, lift), and STM / agents and come to the conclusion the language is heavily in flux

http://news.ycombinator.com/item?id=1150369

http://www.jroller.com/vaclav/entry/secret_agents_in_scala

(and goat rodeo)

- Or you can read the Oreilly and Artima books on the type system, figure out how it's designed for development in the large. And go back and read Guido's ,um, what do you call them, jeremiads or screed or something like that.

> a valid Java program is also a valid Groovy program), so I'm not sure what Extend brings to the table.

Groovy is dynamically typed, not statically typed. Though a month ago Springsource/EMC hired someone to build a static type checker (called Grumpy) for Groovy (http://www.jroller.com/melix/entry/groovy_static_type_checker_status) with the usual solicit for free labor by spinning the "open source" tale.

Presumably they heard about Xtend from the same place Jboss/Redhat and Jetbrains did. Hence the recent avalanche of announcements regarding Ceylon, Kotlin, and Grumpy.

I love Scala, but some deem it too "complex". Java's successor needs to be easy enough to understand for the millions of Java programmers that will have to learn it. Enterprises want a language any code monkey can write in...

http://lamp.epfl.ch/~odersky/blogs/isscalacomplex.html

This is where Ceylon / Fantom are interesting: they improve a little on Java, removing the various language warts, while not being too different.

http://www.jroller.com/scolebourne/entry/the_next_big_jvm_language1

That being said, I'd love it if Scala were the next Blub language ;)

tells you how to use StartSSL, which generates the key in your browser.
Whoops, your private key is now known to another server on this internet

Pardon my ignorance, but I thought the in-browser certificate creation process avoids sending the private key.

https://developer.mozilla.org/en-US/docs/Web/HTML/Element/keygen

http://www.jroller.com/whoami/entry/browser_generated_certificate_requests (2006)