January 2008 - challenge-discuss

Are consultants considered Sun employees?
by David Collier-Brown 31 Jan '08

31 Jan '08

Then official rules say: Employees of Sponsor and Sun Microsystems, Inc. ("Sun"), and each of their respective parent companies, affiliates and subsidiaries, participating advertising and promotion agencies (and members of their immediate family, defined as parents, children, siblings and spouse, regardless of where they reside, and/or those living in the same household) are not eligible. Sun Campus Ambassadors are eligible to participate. I'm a contractor working for Sun: am I excluded? --dave -- David Collier-Brown, | Always do right. This will gratify System Programmer and Author | some people and astonish the rest davecb(a)spamcop.net | -- Mark Twain (416) 223-5943

3 3

So so few/high prizes?
by Clemens Eisserer 30 Jan '08

30 Jan '08

Hi, I've just subscribed to the list, so please excuse if this has been asked already before. What I wonder is why the prizes are so large (25.000$ each) and why there are only so few. I am a student and thought implementing a Java2d-Xrender-backend and some enhancements to the GtkLnf (which I would have done also without the prizes) would be a quite cool project for maybe a 2500$ prize - but this way - I don't think I have any chance at all. There will be a few high-skilled extremly motivated teams, that will work hard to get the money, and when its done, they are away. Another downside is that those teams will not work on "groundwork" - they will work on shiny new stuff leaving which will make everybody saying *wow* - but which won't solve existing problems. But if there would be e.g. 50x3500$ also small projects could participiate, maybe even with some kind of range like 1. is 15000, 2. is 10000, 3. is 7500 .... I am just interested what the idea behind this high prizes is? Why was it choosen in favour of many small one? Do you really think this will lead to improved open-source contributions? Thank you in advance, lg Clemens

2 2

A small idea about java API for URI
by John Hax 30 Jan '08

30 Jan '08

The contest seems very interesting, but I don't know whether I can make a complete proposal. Here is just my story and my idea. Last weekend I was writing my own URI implementation in JavaScript because there is no standard library in JS. (Oh, I'm a web developer not a pure Java guy :) I need to design an API for my URI class, so as usual, I use java API for reference. After one day hard work, I finished a clone of java.net.URI class in JavaScript. But I noticed that the documentation of java.net.URI says there are some deviations from the spec so I decide to read the spec (RFC2396) to check it. To my surprise, I found RFC2396 has been obsoleted by a new standard: RFC3986. It introduces many changes, for example, it doesn't use terms "scheme-specific-part", "opaque URI" any more; the definition of "absolute URI" has been changed (or clarified); some markers (such as ! *) have been moved from unreserved category to reserved category; it also include the clarification of empty path, relative reference, the resolution of the URI reference...All those improvements make java.net.URI outdated and not standard-compliant. The API for URI should be redesigned according to the latest standards, we may introduce a new URI class and mark the old java.net.URI deprecated. We also need some subclasses of URI for normal schemes, such as URN, HttpURI, FtpURI, FileURI...for scheme-specific operations (for example, it will introduce extra rules for equality and normalization), and AbstractURI class for building a new scheme. It also be very nice to implement a superset of URI, that is IRI, as RFC3987. (The docs of .NET System.Uri class says it implements the IRI within Uri class.) Finally, because URIs are just identifiers, and it has been the infrastructure of the most systems, so it may be not proper to put URI classes in java.net package. I think it should be java.util.URI. So, that's my small idea, redesign and implement URI/IRI classes, move them to java.util package and mark the old java.net.URI deprecated.

1 0

How complete should a draft proposal be?
by Wayne Meissner 29 Jan '08

29 Jan '08

Is it ok to post a draft proposal once we have just the outline of the idea, so we can get feedback to see if the idea is viable and/or will meet all the requirements, before going further with the proposal? i.e. is incremental draft proposal development & discussion ok, or should only "almost complete" drafts be posted?

2 1