[XSL-LIST Mailing List Archive Home] [By Thread] [By Date]

Re: [xsl] Replacing = with == and ===

Subject: Re: [xsl] Replacing = with == and ===
From: "Michael Kay mike@xxxxxxxxxxxx" <xsl-list-service@xxxxxxxxxxxxxxxxxxxxxx>
Date: Tue, 5 Aug 2014 07:17:44 -0000

On 4 Aug 2014, at 23:14, L2L 2L emanuelallen@xxxxxxxxxxx
<xsl-list-service@xxxxxxxxxxxxxxxxxxxxxx> wrote:

> ..... Wow how funny is this....
>> everyone seems to expect browser-based technology to be free
> ... Why you--a wizard(meaning experience developer)-- the one making it?
Wouldn't it be better to get people to do it for free? Isn't that what GitHub
for? You have the source code on sourceforge. sourceforge is not like GitHub

Thanks for the suggestion. But I have 15 years experience of developing Saxon
in both open source and commercial versions, I have had many offers of
third-party contributions, and I have yet to get a single one that was tested
and bug-free and well-written and documented. There's third-party contributed
code in there, but in every case, integrating it was as much work as if I had
written it myself. You'd be surprised how many people offer code and then
react with horror when you ask for tests. Let alone asking them to sign
something showing that their employers are happy for their work to be
published under an open source license.

There are of course successful open source projects that rely on crowd-sourced
contributions. But they require very strong central management and quality
assurance. If I had wanted to manage large teams of programmers I would have
stayed in my previous job. When I started out on Saxon it was with a strong
belief that writing software was like writing books or music; you don't get a
good product (and you don't get it fast) if it is done by committee.

Saxon-CE is on GitHub and you are very welcome to "do the GitHub thing": make
a fork, add your own improvements, and offer them back. Be aware, though, that
we won't integrate your improvements unless they meet very high standards.

In the XML space, community development by volunteer users working in their
spare time has conspicuously failed. Look at how libxml and libxslt have
stagnated: they have millions of users, everyone wants improvements, but
no-one has been able to find the time and effort to take the technology
forward once the original developer lost interest (or ran out of money).

Michael Kay

Current Thread