dotgnu-general
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [DotGNU]The ECMA 334 Examples (from spec)


From: James Michael DuPont
Subject: Re: [DotGNU]The ECMA 334 Examples (from spec)
Date: Thu, 24 Oct 2002 07:33:08 -0700 (PDT)

First of all, I want to apologise in advance for my nasty comments
that I am about to make.  Again, I am forced to alienate my fellow
hackers and take a radical stance on the issues. 


If noone else is willing to take a risk and fight for our rights, then
I will have to, with our without your support.

Sorry if you dont aggree with me, but we cannot just keep on running
away from microsoft for no reason at all.

Please remember here, I AM not a lawyer, most of you are not either.

> But I agree , the copy+paste from the spec to test the spec
> is illegal as well ;-)... 
I dont aggree at all. copy+paste+send maybe. copy+paste+send as your
own work, yes. If you redistribute the entire spec in a way that it can
be compiled, that is fine. If you write a program that can parse and
compile the spec as it, even better.

Redistribution of the parts of the ecma spec is
touchy.
There are ways to legally cite parts of the spec, normative and
non-normative. Lots of the spec that are not supposed to be normative
have mistakenly been made normative(see the jaggersoft page).

Also you can cite percentages of the spec in your own documents.

If you have do that, then it different than distributing software that
can run the compiler over pdf files. 

I have to take a completly different approach to this.

Because I am not offically part of the DotGNU project, I can take 
this risk apon myself. 

DotGNu does not want to look bad, so it is my job to look bad. It Does
not bother me at all.

I would love to get sued by MS, it would look great on my resume.

Also, the letter from MS said not to distribute the example separatly.
I wont do that. 

here is the important part :

1. The crashes and bugs that I got from the running the code through
the compiler need to be condensed down into a simple reproducable
example and posted to the bug page. 

2. That has to be done by a person who knows the compiler well and can
condense the problem into a test case. This is creative work that can
be copyrighted. You cannot write a program to do that, only narrow down
the testing. 

3. The descriptions of the crashes, the references to the sources of
the errors is completly legal. The only issue is the redistribution of
parts of the original documents.



Making a testing tool that compiles sources code from a google search
on C# and then reports crashes is completly legal. Using XPDF as a
compiler front end is as well. Calling the ECMA spec my own work is
illegal. Calling the tests that you collect off the web yours,
or even redistributing them could be a violation.

The error message should be :
"From the web page XYZ, from line/column abc to line/column xyz
the code that was parsed there crashed the cscc with the parse stack
A::B::C::D" 
This should be in a format that can be reproduced, and
the even cached. 
The cached data from that page is stored here on this server FGH.
How come no-one sues google for caching thier pages? We can cache all
the data we need for fast redistribution.

Running away from microsoft is not the way to do it. You need to help
the project by being rational, not irrational. You need to present
logical arguments, not voodoo-hoodo-hocus-pocus.

Please just present facts and not fears, I get the feeling that 
you just create more FUD by running away from the problem.

The importance of this issue is really to end the FUD about the 
spec. 

The DOtGNU page have all this crazy stuff about non-published ms docs,
yet offer no guidelines. Most people use the annotated wordfiles from
MS and not the ECMA spec. 

The DotGNU page does not help at all, it just creates more FUD. It has
to stop.

I will be re working my testing program as a general PDF front end to
cscc. You guys can just ignore it if you want. I will be posting the
segfaults and compiler errors to the savannah page.

This will create a clean room testing environment and everyone will
benefit.

mike

=====
James Michael DuPont
http://introspector.sourceforge.net/

__________________________________________________
Do you Yahoo!?
Y! Web Hosting - Let the expert host your web site
http://webhosting.yahoo.com/


reply via email to

[Prev in Thread] Current Thread [Next in Thread]