Repeating my posts for the freelists site

[Harry Binswanger hb@xxxxxxxx]

For Edward and any others trying to read at www.freelists.org, here are my
posts that didn't get through, sent in plain text, which does seem to get
through.

Date: Tue, 10 Apr 2018 08:21:28 -0400
To: xywrite@xxxxxxxx
From: Harry Binswanger
Subject: Re: A radical idea: a new XyWrite
Let me preface this by acknowledging that I don't know enough to have an
accurate judgment of such things, but let me give a counterargument in the
hope that something can be done.
1. The task I'm suggesting is orders of magnitude simpler than producing a
WYSIWIG program (even Eudora is WYSIWYG), let alone a photo-editor.
2. It wouldn't be a commercial product and wouldn't have to have the
quality of XyWrite. Why not? Because in event of a glitch, we have XyWrite
(vDOSPlus) right there to carry on with. It doesn't require abandoning
XyWrite, just having another tool. I would expect a slow migration to a
progressively higher percentage of time spent with the new tool.
3. I wasn't thinking of porting but of cloning. (In the old days, there was
something called cross-compiling, and I wonder if that exists and is relevant.)
4. As to who would pay, I would, for one, and I think several of us would
contribute.
Again, I'm not saying that these counterarguments carry the day. Just that
we should look further.

--Harry


Phil,

Hi! Are you the Savior I've been looking for? :)
By all means, see if steps can be taken. The state of ownership of XyWrite
code is unclear. It appears to be abandonware. But the previous owners, The
Technology Group, are lawyers, so proceed with caution.
What is DOS Macro 11?
Do you think a porting would be easier than just writing a new editor in
C++ or Python? I mean, isn't writing a text editor an assignment given in
computer science classes?
Another question, if you'll be so kind. The fallback position is just to
break the 64k memory barrier (then use vDOSPlus). Am I right that it would
be fairly simple to plug into the existing code a swap of pages in and out
of memory to accomplish this?

Regards,
Harry

Date: Tue, 10 Apr 2018 08:41:09 -0400
To: xywrite@xxxxxxxx
From: Harry Binswanger
Subject: RE: A radical idea: a new XyWrite

So, it seems that there are 3 ideas that we should keep distinct.

1. Modifying the existing code, such as adding a page-swap of memory.
2. Translating the existing code into a Windows-friendly product. (Porting
or something like that.)
3. Ignoring the existing code and writing, from scratch, a new program that
will a) give us, the users, the same functionality as XyWrite, or with
trivial differences, and b) be able to use, or automatically translate, all
the nonASCII codes of XyWrite and run XPL.
I suspect that there would be no issue of property rights involved here,
since what we'd be doing is using a different means to accomplishing the
same end as XyWrite does. And anyway, no harm, no foul--and who is being
harmed by us 36 guys using a new program to do the same things as an old
program that has been abandoned for 12 years or more?

Which of these three alternatives is best?

Regards,
Harry

Date: Tue, 10 Apr 2018 16:15:36 -0400
To: xywrite@xxxxxxxx
From: Harry Binswanger
Subject: Re: A radical idea: a new XyWrite

Kari Diabolus,

Greetings. Here is the other, other side.
> 1. It is way too late to do anything of the sort. The customer base which
> still exists is too small (and dying one by one).
I will pay thousands of dollars to get this. I suspect others would pay a
considerable, if lesser amount. What matters it if our numbers are few and
dwindling? We aren't trying to please any set of people but ourselves.
> 2. We already have a product that is ported to the modern Windows
> platform, albeit a singular one (Nota Bene).
I have tried NB with each iteration, without success. I don't know that
I've tried NB 3, which you have mentioned, but certainly most of them for a
long, long time. NB is indeed sufficiently XyWrite-like, but it is terribly
slow and non-robust. Versions later than NB 9 will not load U2. A few years
back, I wrote a post to this list on how NB would save us, but despite much
tinkering with each version and some real improvement by means of that
tinkering, I have always found NB unsatisfactory for scripting work, which
is most of what I use XyWrite for.
(You may not know that I developed a front-end language that I call XyBasic
to make scripting more intuitive. So far, it has an installed user-base of
one; but it's been only 28 years since I offered it to the world.)
> 3. Harry and Carl and many others would like to see a XyWrite 4
> derivative, whereas a XyWrite 3-type program would require less work and
> be more likely negotiable with any parties holding the rights. This would
> make the project less desirable to key proponents but more feasible on the
> whole.
That's possible, except that one of the very best features is available
only in Xy4: wildcard search and replace. I use it all the time. (Carl has
made regex available to us via U2, and that conceivably could substitute.)
I don't think the rights are at stake any longer. How long has it been
since TTG even was contactable? 12 years? 15 years?
> 4. vDosPlus XyWrite is quite good as is.
It is indeed. But it still has the 64k memory limit, and it runs 15 times
slower than XyWrite. What difference to that make, you ask? I have large
programs that I not only need to run but, more importantly, need to compile
from XyBasic. In Xy4, the compilation (which is partly done by an external
program but partly by XyWrite itself) can take over a minute. Since I am a
clumsy programmer, I need to recompile maybe 50 times a day, when I'm
developing code, so that's a deal-breaker.
(The workaround for the latter problem is for me to write a Python program
to do compiles, but this would take me maybe 400 hours. Happy hours, true.
But still . . .)
> 5. Editing huge XyWrite files with aplomb can be accomplished using
> EditPad with syntax highlighting of Xy codes.
My own problem is not filesize, and Carl's Big Edit frames for U2 take care
of that anyway. I have EditPad but my main alternative is Notepad++, which
also has syntax highlighting for whatever. It's good, but not as good as a
command-line, easily customizable program--and one with which I have 31
years of familiarity.

And lastly, as de Gaulle put it: vaste programme, monsieur!

Well, you know de Gaulle is no longer trending.

Regards,
Harry

Date: Tue, 10 Apr 2018 16:19:32 -0400
To: xywrite@xxxxxxxx
From: Harry Binswanger
Subject: RE: A radical idea: a new XyWrite

Phil,

Of course, I know what DOS is. I don't know the Macro 11 part. No matter.
On conversion of code: Xy3+ is written in Assembler. Are you saying that
there's some automated conversion of that into a modern language?!
If Xy4, I just learn is written in some flavor of C, perhaps we could buy
the rights to use the source code (not the binary) for the purpose of
converting it.

This is hopeful!

Regards,
Harry