The Linux problem, and a suggestion

[Peter Alcibiades]

We have been brought face to face with this again on the issue of font 
support.  Coincidentally the other day an old friend asked me whether he 
should take up Rev or Python.  I have a spare Rev license and offered it to 
him, and we had a discussion.

If Linux is your main system,  you do have choice.  Python with its various 
gui IDEs is a real alternative.  Now, there are a lot of people on this 
list who feel and express disdain for Python, often on the basis of limited 
experience of earlier versions of some of the IDEs.  But lets be clear:  
its a viable alternative.  It works.  You don't have font problems.  You 
can print.  You have bindings for any OSS database you want.  Multiple 
desktops just work.  You have an enormous choice of editors for your code, 
you can use anything from Gedit to the (to me) vastly overfeatured Eclipse, 
stopping off at Vi or Emacs on the way.  It has a track record in large and 
small scale projects.  And its free.

The problem with Python as compared to Rev is that you end up writing more 
code, and arguably more complicated code.  That's what everyone on the list 
will say.  And if its Windows or Mac, they will be right.  If its Linux 
they will be plain wrong.  If you end up writing chunks of your stuff in 
Awk, because you cannot get it to work properly in Rev, you are not writing 
less or simpler code.  And it is not easily cross platform either.

So what is to be done?

The first thing is to decide, is there a viable business for Rev in the 
Linux programming environment market?  It is not clear.  The presence of so 
many good free tools, and the predominance of Perl and Python must make it 
hard.  Not to mention that you also have Lua, Ruby and so on.  On the other 
hand, Rev has some unique advantages.  It really is faster and simpler if 
it would only work as well in Linux as it does on the other platforms.  But 
we will never find out if there is a viable business, going on as we are.  
Because no rational professional will commit himself to Revolution on Linux 
as his principal programming tool.  It makes no sense, with the level of 
infelicities even an amateur can find.  And the pros find more than I do!

Suggestion: it will not take a huge investment to clean them up, and 
that we could then find out one way or the other, and make a decision.  If 
its viable, or shows signs of increased sales and use, carry on.  If not, 
drop it and put the efforts to better use.  There is one reason why it may 
be worth doing.  That is, Linux is a tiny proportion of the market, but it 
is disproportionately developers and technical people.   So market share in 
Linux can be worth a great deal, indrectly, because it gets to a key market 
segment.

If you were going to do this, what would you do?  In one sentence, make it 
a well behaved Linux application.  Right now, it ain't.

Fix printing.  It must be possible to just print fields using the standard 
commands.  You should not have to go out to shell and invoke Awk and pipe 
through to kprinter just to print a field.

Fix multiple desktops.  Rev is the only Linux package I know that doesn't 
support them properly.  Why?  Why is it that I have 6 virtual desktops and 
am obliged, in Rev alone, to have dictionary, editor, app itself, property 
inspector, palettes, all crammed into one of them, one under the other 
cluttering up the one desktop, when any other app would let me split them 
up.  It was intolerable with an 18 inch screen, and its only a bit better 
with a 22 inch wide screen.  And its such basic stuff, this!

Fix fonts.  I have no idea how Rev picks what fonts it can and cannot see, 
and we get no answers, but every other Linux application there is can 
handle fonts in Linux.  Why not Rev?

Fix the editor.  You must have some way of using the standard Linux 
programming editors in an integrated way.  Do plugins for Geany and Kate at 
least.  Maybe there are others.  I guess some will want Eclipse.  When you 
have all the super duper functionality of the IDE only available through a 
third party addon, tRev, which does not come for Linux, what you are doing 
is making the Linux version a seccond class citizen.  You don't have to 
make tRev available, but you do have to find some way to offer similar add 
on functionality.

Fix the size issue.  It must be possible to SEE the palette of controls and 
the dictionary.  Every other Linux application manages it, regardless of 
monitor size.  Why is Rev on a 22 inch screen showing up in what looks like 
4 point font, with no way of changing it?  Am I really supposed to peer at 
my screen through reading glasses, just to use Rev?

If you will not or cannot fix the size and desktops issues right away, at 
least make the dictionary open in a browser window.  That way we can resize 
it ourselves, and we can also have it open on a different desktop.

Deliver the Player.  The Media Player has been coming soon to Linux for 
years.  It may not be essential as a feature, but if its there for Windows 
and Mac, it has to be there for Linux.  

Finally there is Rev Browser.  I bought or maybe was given this when it was 
altBrowser, and have been waiting patiently for it ever since.  It may not 
be essential functionality in itself, but if its on Windows and Mac, it 
must be available on Linux.

So, how did the conversation which started this go?  I offered my free 
license.  I pointed him to this list and told him what nice people we all 
are.  I explained that Rev for Linux is not the same as Rev for the other 
platforms, and how and why that is.  I also gave him a bunch of pointers to 
material on Python.

Sad to say, they were both free, but he declined the Rev license.  He was 
very diplomatic.  He said he didn't think he would make good use of it.  
And he is buying Mark Lutz' O'Reilly book.  I wanted to tell him he was 
making a mistake.  But I couldn't.  What I would really like is for the 
next time I have this conversation, to just be able to say, do it, its 
fully cross platform, you'll never look back.

Well?  How about it?

Peter