Re: we now have "lilypond" organization on GitHub

[David Kastrup]

Joseph Rushton Wakeling <address@hidden> writes:

> On 24/09/13 15:41, David Kastrup wrote:
>> What about "Try it" did you not understand?  Windows does not just allow
>> you to say
>>
>>      sudo apt-get build-dep lilypond
>>
>> Instead you have several dozens of dependencies you have to satisfy by
>> hand, and then the fun with registry entries and other stuff starts.
>
> I didn't need to try it -- I already anticipated that it would involve
> huge amounts of hassle setting up the dependencies.  I asked you
> simply because it was important to me to understand the problem as
> _you_ saw it.

No, you are not just "asking".  You are throwing diagnoses around and
proposing solutions that are known not to work.

>> No doubt about that, but we're not in the situation to fix Windows.
>
> Is the problem Windows, or that many of the GNU dependencies are
> difficult to install on Windows?

That's the same problem.

> I don't know what kind of issues you have in practice with debugging
> and diagnosing Windows-related issues.  Given other factors I can
> imagine that it might be less of a problem than the hassle involved in
> setting up development on Windows.
>
> Still, I'm curious -- what is it about making a Windows development
> environment for Lilypond that doesn't make sense?  Is it just the
> hassle with the dependencies, or are there other factors?

You are probably assuming that Windows is the market leader for desktop
operating systems because it the choice of engineers, picking the best
product made by the best engineers to be found.  Or at least a
reasonable approximation: it can't be thoroughly rotten if it's the
market leading product.

It's a reasonable assumption to make for market-leading computer
products.

Now it's not just Microsoft, but I've rather made the experience that
pretty much _every_ market leading software product where an innocent
bystander would think "they made the basic right choices and have a
coherent design and sane interfaces, or they would not be where they are
now" tends to be a steaming heap of fetid dingo feces.

Microsoft must have acquired the _gods_ of marketing.  But engineering?
No.  If they have, their work does not survive the processes.

_Don't_ _ever_ make assumptions about how things must be possible or
easy natively under Windows since its position as the leading desktop
operating system implies that it must be a reasonably workable system
with reasonably complete and/or working facilities.

> My response to Phil wasn't meant to be cheap pontification, it was
> meant to be simply: "Here are a list of reasons why you shouldn't be
> complacent about the usability of your tools."

It's always fun to suggest eating humble pie to others.  But I really
recommend getting some experience _before_ lecturing them.

-- 
David Kastrup