Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.forth > #57836 > unrolled thread

Dropbox rant

Back to article view | Back to comp.lang.forth

Contents

  Dropbox rant Mark Wills <markwills1970@gmail.com> - 2017-04-22 11:24 -0700
    Re: Dropbox rant Cecil Bayona <cbayona@cbayona.com> - 2017-04-22 13:50 -0500
      Re: Dropbox rant Andrew Haley <andrew29@littlepinkcloud.invalid> - 2017-04-23 04:50 -0500
        Re: Dropbox rant Julian Fondren <julian.fondren@gmail.com> - 2017-04-23 06:58 -0700
          Re: Dropbox rant foxaudioresearch@gmail.com - 2017-04-24 18:06 -0700
            Re: Dropbox rant Julian Fondren <julian.fondren@gmail.com> - 2017-04-24 20:30 -0700
              Re: Dropbox rant foxaudioresearch@gmail.com - 2017-04-24 22:02 -0700
                Re: Dropbox rant "Elizabeth D. Rather" <erather@forth.com> - 2017-04-24 21:52 -1000
                Re: Dropbox rant Julian Fondren <julian.fondren@gmail.com> - 2017-04-25 16:38 -0700
                  Re: Dropbox rant hughaguilar96@gmail.com - 2017-04-25 16:55 -0700
              Re: Dropbox rant Julian Fondren <julian.fondren@gmail.com> - 2017-05-28 17:50 -0700
            Re: Dropbox rant "Elizabeth D. Rather" <erather@forth.com> - 2017-04-24 21:26 -1000
              Re: Dropbox rant john <anon@example.com> - 2017-04-25 10:22 +0100
                Re: Dropbox rant lehs <skydda.barnen@gmail.com> - 2017-04-25 04:14 -0700
                Re: Dropbox rant hughaguilar96@gmail.com - 2017-04-25 11:25 -0700
                  Re: Dropbox rant Alex <alex@rivadpm.com> - 2017-04-26 15:57 +0100
                    Re: Dropbox rant Ron Aaron <rambamist@gmail.com> - 2017-04-26 18:41 +0300
                      Re: Dropbox rant Alex <alex@rivadpm.com> - 2017-04-26 17:59 +0100
                        Re: Dropbox rant Ron Aaron <rambamist@gmail.com> - 2017-04-26 21:38 +0300
                          Re: Dropbox rant rickman <gnuarm@gmail.com> - 2017-04-26 17:48 -0400
                      Re: Dropbox rant hughaguilar96@gmail.com - 2017-04-26 12:42 -0700
                        Re: Dropbox rant Ron Aaron <rambamist@gmail.com> - 2017-04-27 06:57 +0300
                          Re: Dropbox rant hughaguilar96@gmail.com - 2017-04-26 21:20 -0700
                            OT: Hugh's in need of help (was Re: Dropbox rant) Ron Aaron <rambamist@gmail.com> - 2017-04-27 07:45 +0300
                              Re: OT: Hugh's in need of help (was Re: Dropbox rant) rickman <gnuarm@gmail.com> - 2017-04-27 01:31 -0400
                                Re: OT: Hugh's in need of help (was Re: Dropbox rant) Ron Aaron <rambamist@gmail.com> - 2017-04-27 08:43 +0300
                                  Re: OT: Hugh's in need of help (was Re: Dropbox rant) rickman <gnuarm@gmail.com> - 2017-04-27 01:56 -0400
                                  Re: OT: Hugh's in need of help (was Re: Dropbox rant) hughaguilar96@gmail.com - 2017-04-27 13:17 -0700
                                    Re: OT: Hugh's in need of help (was Re: Dropbox rant) rickman <gnuarm@gmail.com> - 2017-04-27 16:31 -0400
                                    Re: OT: Hugh's in need of help (was Re: Dropbox rant) Ron Aaron <rambamist@gmail.com> - 2017-04-28 07:48 +0300
                                  Re: OT: Hugh's in need of help (was Re: Dropbox rant) Ron Aaron <rambamist@gmail.com> - 2017-04-28 10:22 +0300
                                    Re: OT: Hugh's in need of help (was Re: Dropbox rant) "HAA" <someone@microsoft.com> - 2017-04-30 11:09 +1000
                                      Re: OT: Hugh's in need of help (was Re: Dropbox rant) hughaguilar96@gmail.com - 2017-04-29 21:52 -0700
                                        Re: OT: Hugh's in need of help (was Re: Dropbox rant) Ron Aaron <rambamist@gmail.com> - 2017-04-30 08:24 +0300
                                          Re: OT: Hugh's in need of help (was Re: Dropbox rant) rickman <gnuarm@gmail.com> - 2017-04-30 01:54 -0400
                                            Re: OT: Hugh's in need of help (was Re: Dropbox rant) Ron Aaron <rambamist@gmail.com> - 2017-04-30 09:04 +0300
                                              Re: OT: Hugh's in need of help (was Re: Dropbox rant) rickman <gnuarm@gmail.com> - 2017-04-30 02:19 -0400
                                                Re: OT: Hugh's in need of help (was Re: Dropbox rant) lehs <skydda.barnen@gmail.com> - 2017-04-30 00:41 -0700
                                                  Re: OT: Hugh's in need of help (was Re: Dropbox rant) Ron Aaron <rambamist@gmail.com> - 2017-04-30 11:01 +0300
                                                    Re: OT: Hugh's in need of help (was Re: Dropbox rant) lehs <skydda.barnen@gmail.com> - 2017-04-30 01:24 -0700
                                                    Re: OT: Hugh's in need of help (was Re: Dropbox rant) Mark Wills <markwills1970@gmail.com> - 2017-04-30 03:33 -0700
                                                      Re: OT: Hugh's in need of help (was Re: Dropbox rant) jenfhaomndgfwutc@murmic.plus.com - 2017-04-30 16:19 +0100
                                                      Re: OT: Hugh's in need of help (was Re: Dropbox rant) Cecil Bayona <cbayona@cbayona.com> - 2017-04-30 16:46 -0500
                                                      Re: Dropbox rant "HAA" <someone@microsoft.com> - 2017-05-02 11:33 +1000
                                      Back on-topic: what attracted you to Forth?' (was: OT: Hugh's in need of help ...) Ron Aaron <rambamist@gmail.com> - 2017-04-30 08:42 +0300

Page 1 of 3  [1] 2 3  Next page →

#57836 — Dropbox rant

<rant>
How, in the name of all that is holy, can Dropbox.exe
need 98,540 *KILOBYTES* of RAM to do its thing?

It's not much more than a glorified FTP Client/Server.

FFS.
</rant>

[toc] | [next] | [standalone]

#57838

On 4/22/2017 1:24 PM, Mark Wills wrote:
> <rant>
> How, in the name of all that is holy, can Dropbox.exe
> need 98,540 *KILOBYTES* of RAM to do its thing?
> 
> It's not much more than a glorified FTP Client/Server.
> 
> FFS.
> </rant>
> 

That is an illustration of why Forth is not popular, there is absolutely 
no interest  or even knowledge of what being efficient is.

Forth is a tool set for those who want to efficiently use the machine 
resources, most programmers today have no idea of the concept, megabytes 
upon megabytes to do the simplest task.

My favorite Forth for a very long time was Pygmy Forth, 19Kb, with 
Interpreter, Compiler, De-compiler, Assembler, Disassembler,  Meta 
Compiler, and an editor.

-- 
Cecil - k5nwa

[toc] | [prev] | [next] | [standalone]

#57896

Cecil Bayona <cbayona@cbayona.com> wrote:
> On 4/22/2017 1:24 PM, Mark Wills wrote:
>> <rant>
>> How, in the name of all that is holy, can Dropbox.exe
>> need 98,540 *KILOBYTES* of RAM to do its thing?
>> 
>> It's not much more than a glorified FTP Client/Server.
>> 
>> FFS.
>> </rant>
> 
> That is an illustration of why Forth is not popular, there is absolutely 
> no interest  or even knowledge of what being efficient is.

The times they are a changin'.  If you run servers in the cloud you
pay for processors and you pay for bytes, so there is a considerable
renewed interest in fast and compact systems and protocols.

Andrew.

[toc] | [prev] | [next] | [standalone]

#57908

On Sunday, April 23, 2017 at 4:50:31 AM UTC-5, Andrew Haley wrote:
> 
> The times they are a changin'.  If you run servers in the cloud you
> pay for processors and you pay for bytes, so there is a considerable
> renewed interest in fast and compact systems and protocols.
> 
> Andrew.

I have a little script in Perl that continuously makes HTTP requests.
It's actual function is to serialize internal requests to an
application that tends to fall over if it gets asked 20 expensive
questions at once. The script itself doesn't actually accept
HTTP requests. It's one half of the solution. The other half is a CGI
script that either says "oh yeah here's your (cached) answer
immediately" or "you want what? never heard of it. enjoy your error.
(psst-- they want this new thing now. I've added it to the list!)"

This little script grows by ONE MEGABYTE each request. Permanently. I
have no idea why. I tried switching out LWP, I tried not checking for
the JSON-validity of the data it was fetching. The only other thing
the script does is log! In the end I resolved things well enough (that
I can't now justify spending any more time on the matter) by having it
kill itself after it bloats past 200MB.

This was a server with 8GB of memory. This was in 2017. This was with
a relatively nice high level language. And the entire time I was
trying to find out what was going on, I was thinking, maaan, if only
this were in Forth...

In Forth I would've tested the JSON with a streaming validator working
in fixed memory, writing to a file as fast as I got more JSON from the
remote server. It's inconceivable that I'd find myself worrying about
a similar Forth program taking 7.1G of RAM, or that I'd have any
trouble pinning down where that came from.

[toc] | [prev] | [next] | [standalone]

#58041

On Sunday, April 23, 2017 at 9:58:03 AM UTC-4, Julian Fondren wrote:
> On Sunday, April 23, 2017 at 4:50:31 AM UTC-5, Andrew Haley wrote:
> > 
> > The times they are a changin'.  If you run servers in the cloud you
> > pay for processors and you pay for bytes, so there is a considerable
> > renewed interest in fast and compact systems and protocols.
> > 
> > Andrew.
> 
> I have a little script in Perl that continuously makes HTTP requests.
> It's actual function is to serialize internal requests to an
> application that tends to fall over if it gets asked 20 expensive
> questions at once. The script itself doesn't actually accept
> HTTP requests. It's one half of the solution. The other half is a CGI
> script that either says "oh yeah here's your (cached) answer
> immediately" or "you want what? never heard of it. enjoy your error.
> (psst-- they want this new thing now. I've added it to the list!)"
> 
> This little script grows by ONE MEGABYTE each request. Permanently. I
> have no idea why. I tried switching out LWP, I tried not checking for
> the JSON-validity of the data it was fetching. The only other thing
> the script does is log! In the end I resolved things well enough (that
> I can't now justify spending any more time on the matter) by having it
> kill itself after it bloats past 200MB.
> 
> This was a server with 8GB of memory. This was in 2017. This was with
> a relatively nice high level language. And the entire time I was
> trying to find out what was going on, I was thinking, maaan, if only
> this were in Forth...
> 
> In Forth I would've tested the JSON with a streaming validator working
> in fixed memory, writing to a file as fast as I got more JSON from the
> remote server. It's inconceivable that I'd find myself worrying about
> a similar Forth program taking 7.1G of RAM, or that I'd have any
> trouble pinning down where that came from.

What would you say has prevented Forth from being applied to this problem 
domain?

My own prejudiced opinion is that a raw Forth solution would be in-
comprehensible to the majority of IT people, but some kind of application 
specific language tailored to the problem could be made palatable I suspect and
also be efficient.

However most Forth professionals are busy solving a completely different set of
problems, so perhaps there is just not the critical mass of Forth people to 
to get activity in these other areas. (?)

B 

[toc] | [prev] | [next] | [standalone]

#58052

On Monday, April 24, 2017 at 8:06:07 PM UTC-5, foxaudio...@gmail.com wrote:
> 
> What would you say has prevented Forth from being applied to this problem 
> domain?
> 

What prevented me from applying Forth in this exact instance was
simply that the tools didn't exist, and that I couldn't justifiably
write them instead of use tools that did exist.

The tools aren't much:
1. make HTTP requests over TLS
2. validate JSON

JSON can validated one character at a time with a state machine.

HTTP requests are trivial, but I don't know anything about TLS;
you could probably use libcurl.

> My own prejudiced opinion is that a raw Forth solution would be in-
> comprehensible to the majority of IT people, but some kind of application 
> specific language tailored to the problem could be made palatable I suspect and
> also be efficient.

Replace one thing nobody has any skill in with another thing that
nobody has any skill in? The result still loses to what people already
have skill in. Just focus on making a good 'raw Forth solution'.
Usually that involves a measure of 'application specific language'.

> However most Forth professionals are busy solving a completely different set of
> problems, so perhaps there is just not the critical mass of Forth people to 
> to get activity in these other areas. (?)

This kind of talk is why I provided a specific answer your general
question above. If the problem is "tools don't exist", you can just
create them.

[toc] | [prev] | [next] | [standalone]

#58061

On Monday, April 24, 2017 at 11:30:22 PM UTC-4, Julian Fondren wrote:
> 
> This kind of talk is why I provided a specific answer your general
> question above. If the problem is "tools don't exist", you can just
> create them.

I am not sure what "this kind of talk" means exactly Julian. I was not being
antagonistic. 

"Just create them" is a common Forth mantra, but your case demonstrates that
time is finite and we can't do everything. I was wondering why, if it can be 
better in Forth, have the Forth programmers of the world not picked up the 
challenge. I submitted that they are busy doing other kinds of work.

My implication is that one would create an intermediate word set on top of 
Forth that would let others work in the problem space with less mental 
friction. The word names chosen would be familiar to web developer for example. 
Then the Forth based solution would accessible to the people in the field 
and maybe even preferred because of the benefits.

Where my thoughts come from:

Without a critical mass of the world using Forth solutions we are back to the limitations of lifespan for the small group of Forth coders who exist.

It limits how deep Forth and the Forth philosophy can get into the bigger world.

The observation that I see are the big revolutions in computing in the last 20 
years (internet, mobile phones) but the world seems to have forgotten Forth or 
if they know about it, the benefits that we perceive have not been communicated.

Perhaps it is not possible.
It could be that it's just to hard for people to change their paradigms.

Just things that I think about.

B

[toc] | [prev] | [next] | [standalone]

#58068

On 4/24/17 7:02 PM, foxaudioresearch@gmail.com wrote:
> On Monday, April 24, 2017 at 11:30:22 PM UTC-4, Julian Fondren wrote:
>>
>> This kind of talk is why I provided a specific answer your general
>> question above. If the problem is "tools don't exist", you can just
>> create them.
>
> I am not sure what "this kind of talk" means exactly Julian. I was not being
> antagonistic.
>
> "Just create them" is a common Forth mantra, but your case demonstrates that
> time is finite and we can't do everything. I was wondering why, if it can be
> better in Forth, have the Forth programmers of the world not picked up the
> challenge. I submitted that they are busy doing other kinds of work.

Forth programmers don't "just create" tools on spec. Any Forth 
programmer working in a particular application domain develops the tools 
needed, and they remain in his/her toolbox. The interesting thing is, 
that I have seen good Forth programmers tackle brand new application 
domains, and they will develop the needed intermediate tools and 
complete the application in less time than professional C programmers 
have estimated, even with whatever tools they had handy.

> My implication is that one would create an intermediate word set on top of
> Forth that would let others work in the problem space with less mental
> friction. The word names chosen would be familiar to web developer for example.
> Then the Forth based solution would accessible to the people in the field
> and maybe even preferred because of the benefits.

Yes, that is the process we all use. But those word sets would spread 
if, and only if, that code was documented and published for others 
working in that domain. This is where the model breaks down: the 
organization funding the original work either deliberately decides to 
keep the code proprietary to maintain their competitive edge, or is 
reluctant to spend the additional funds to put the code into publishable 
form.

A lot of the libraries in C and other languages are developed either in 
academia, where publishing is expected and supported, or by individual 
programmers on their own time. Academia in general has not embraced 
Forth (see below). The latter case tends not to happen because those 
individual programmers would rather implement a new Forth than develop, 
document, and publish application oriented tools.

> Where my thoughts come from:
>
> Without a critical mass of the world using Forth solutions we are back to the limitations of lifespan for the small group of Forth coders who exist.

Each one of you needs to be developing and cherishing newbies by showing 
them how to use the Forth tools you have developed.

> It limits how deep Forth and the Forth philosophy can get into the bigger world.

Yes.

> The observation that I see are the big revolutions in computing in the last 20
> years (internet, mobile phones) but the world seems to have forgotten Forth or
> if they know about it, the benefits that we perceive have not been communicated.
>
> Perhaps it is not possible.
> It could be that it's just to hard for people to change their paradigms.

People follow the paradigms they learn in college. Back in the day, we 
tried very hard to interest universities in Forth. Some engineering 
depts loved it, because of the productivity, but gave it up because 
their students couldn't get Forth jobs when they graduated. Computer 
science depts hated Forth, because it was too simple.

> Just things that I think about.

As do I.

Cheers,
Elizabeth

-- 
Elizabeth D. Rather
FORTH, Inc.
6080 Center Drive, Suite 600
Los Angeles, CA  90045
USA

[toc] | [prev] | [next] | [standalone]

#58162

On Tuesday, April 25, 2017 at 12:02:40 AM UTC-5, foxaudio...@gmail.com wrote:
> On Monday, April 24, 2017 at 11:30:22 PM UTC-4, Julian Fondren wrote:
> > 
> > This kind of talk is why I provided a specific answer your general
> > question above. If the problem is "tools don't exist", you can just
> > create them.
> 
> I am not sure what "this kind of talk" means exactly

You drive through one neighborhood and are pleased to see that it's
very clean and pretty. You keep driving and then enter a neighborhood
where there's litter everywhere, used diapers by the streetside,
graffiti on street signs and building signs, etc.

If you wonder at the difference, there are immediately two kinds of
questions you can ask: "why don't the people in this second
neighborhood keep it clean?" and "what is wrong with the various
fundamental constants of the universe (planck, etc.), the nature of
humanity, the family structure, post-hunter-gatherer civilization,
Western legal traditions, global climate trends, preschool
education, and the GDP - that result in the cleanliness privilege
some neighborhoods and the cleanliness deprivation of others?"

The first question regards a specific population in a specific
circumstance, and their specific skills and agency (not littering;
cleaning up litter). It can at least potentially be directly answered
by the people involved: each individual who produces some of the mess
or fails to clean it up can explain themselves. If you wanted to
improve things it's obvious how you should proceed: discourage people
from littering, encourage people to pick up litter, and actually clean
things up yourself. Even if advocacy is not one of your skills, you
can still directly contribute to the cleanliness of the neighborhood.

The second question is fun to try and answer. And maybe you'll come up
with a really plausible-sounding reason as to why the phase of the
moon uniquely affects this one neighborhood and not the other. But no
matter how plausible-sounding the reason (lets not pretend that truth
is an option here), it won't be one that you can do anything about.
You can't directly contribute to the correction of the moon, and no
matter how sweetly you advocate that the moon be corrected, who is
going to listen to you who can directly contribute to fixing it? The
entire discussion is only good for passing the time.

There's the drunkard's search:

  A policeman sees a drunk man searching for something under a
  streetlight and asks what the drunk has lost. He says he lost his
  keys and they both look under the streetlight together. After a few
  minutes the policeman asks if he is sure he lost them here, and the
  drunk replies, no, and that he lost them in the park. The policeman
  asks why he is searching here, and the drunk replies, "this is where
  the light is."

And there's the intellectual's search, where you make sure to look
for your keys only where they would be irretrievable if found.

[toc] | [prev] | [next] | [standalone]

#58163

On Tuesday, April 25, 2017 at 4:38:30 PM UTC-7, Julian Fondren wrote:
> On Tuesday, April 25, 2017 at 12:02:40 AM UTC-5, foxaudio...@gmail.com wrote:
> > On Monday, April 24, 2017 at 11:30:22 PM UTC-4, Julian Fondren wrote:
> > > 
> > > This kind of talk is why I provided a specific answer your general
> > > question above. If the problem is "tools don't exist", you can just
> > > create them.
> > 
> > I am not sure what "this kind of talk" means exactly
> 
> You drive through one neighborhood and are pleased to see that it's
> very clean and pretty. You keep driving and then enter a neighborhood
> where there's litter everywhere, used diapers by the streetside,
> graffiti on street signs and building signs, etc.
> 
> If you wonder at the difference, there are immediately two kinds of
> questions you can ask: "why don't the people in this second
> neighborhood keep it clean?" and "what is wrong with the various
> fundamental constants of the universe (planck, etc.), the nature of
> humanity, the family structure, post-hunter-gatherer civilization,
> Western legal traditions, global climate trends, preschool
> education, and the GDP - that result in the cleanliness privilege
> some neighborhoods and the cleanliness deprivation of others?"
> 
> The first question regards a specific population in a specific
> circumstance, and their specific skills and agency (not littering;
> cleaning up litter). It can at least potentially be directly answered
> by the people involved: each individual who produces some of the mess
> or fails to clean it up can explain themselves. If you wanted to
> improve things it's obvious how you should proceed: discourage people
> from littering, encourage people to pick up litter, and actually clean
> things up yourself. Even if advocacy is not one of your skills, you
> can still directly contribute to the cleanliness of the neighborhood.
> 
> The second question is fun to try and answer. And maybe you'll come up
> with a really plausible-sounding reason as to why the phase of the
> moon uniquely affects this one neighborhood and not the other. But no
> matter how plausible-sounding the reason (lets not pretend that truth
> is an option here), it won't be one that you can do anything about.
> You can't directly contribute to the correction of the moon, and no
> matter how sweetly you advocate that the moon be corrected, who is
> going to listen to you who can directly contribute to fixing it? The
> entire discussion is only good for passing the time.
> 
> There's the drunkard's search:
> 
>   A policeman sees a drunk man searching for something under a
>   streetlight and asks what the drunk has lost. He says he lost his
>   keys and they both look under the streetlight together. After a few
>   minutes the policeman asks if he is sure he lost them here, and the
>   drunk replies, no, and that he lost them in the park. The policeman
>   asks why he is searching here, and the drunk replies, "this is where
>   the light is."
> 
> And there's the intellectual's search, where you make sure to look
> for your keys only where they would be irretrievable if found.

I see some posts from Julien Fondren in which he is making death threats against people, and other posts in which he is just spewing nonsense --- I prefer the latter --- either way though, he makes the ANS-Forth cult look foolish, because they embrace anybody who is loyal, including mentally-ill metaphor-stretching clowns such as himself.

[toc] | [prev] | [next] | [standalone]

#59994

On Monday, April 24, 2017 at 10:30:22 PM UTC-5, Julian Fondren wrote:
> On Monday, April 24, 2017 at 8:06:07 PM UTC-5, foxaudio...@gmail.com wrote:
> > 
> > What would you say has prevented Forth from being applied to this problem 
> > domain?
> > 
> 
> What prevented me from applying Forth in this exact instance was
> simply that the tools didn't exist, and that I couldn't justifiably
> write them instead of use tools that did exist.
> 
> The tools aren't much:
> 1. make HTTP requests over TLS
> 2. validate JSON
> 
> JSON can validated one character at a time with a state machine.
> 
> HTTP requests are trivial, but I don't know anything about TLS;
> you could probably use libcurl.
> 

Yep, libcurl works and quite easily.

https://bitbucket.org/demonview/forth-libcurl/overview
^ includes a 'refresher' that's very close to the one discussed.

[toc] | [prev] | [next] | [standalone]

#58065

On 4/24/17 3:06 PM, foxaudioresearch@gmail.com wrote:
> On Sunday, April 23, 2017 at 9:58:03 AM UTC-4, Julian Fondren wrote:
>> On Sunday, April 23, 2017 at 4:50:31 AM UTC-5, Andrew Haley wrote:
>>>
>>> The times they are a changin'.  If you run servers in the cloud you
>>> pay for processors and you pay for bytes, so there is a considerable
>>> renewed interest in fast and compact systems and protocols.
>>>
>>> Andrew.
>>
>> I have a little script in Perl that continuously makes HTTP requests.
>> It's actual function is to serialize internal requests to an
>> application that tends to fall over if it gets asked 20 expensive
>> questions at once. The script itself doesn't actually accept
>> HTTP requests. It's one half of the solution. The other half is a CGI
>> script that either says "oh yeah here's your (cached) answer
>> immediately" or "you want what? never heard of it. enjoy your error.
>> (psst-- they want this new thing now. I've added it to the list!)"
>>
>> This little script grows by ONE MEGABYTE each request. Permanently. I
>> have no idea why. I tried switching out LWP, I tried not checking for
>> the JSON-validity of the data it was fetching. The only other thing
>> the script does is log! In the end I resolved things well enough (that
>> I can't now justify spending any more time on the matter) by having it
>> kill itself after it bloats past 200MB.
>>
>> This was a server with 8GB of memory. This was in 2017. This was with
>> a relatively nice high level language. And the entire time I was
>> trying to find out what was going on, I was thinking, maaan, if only
>> this were in Forth...
>>
>> In Forth I would've tested the JSON with a streaming validator working
>> in fixed memory, writing to a file as fast as I got more JSON from the
>> remote server. It's inconceivable that I'd find myself worrying about
>> a similar Forth program taking 7.1G of RAM, or that I'd have any
>> trouble pinning down where that came from.
>
> What would you say has prevented Forth from being applied to this problem
> domain?

The fact that no of the organizations that were using Forth were working 
in this problem domain, and the organizations that were working on it 
didn't know about Forth.

> My own prejudiced opinion is that a raw Forth solution would be in-
> comprehensible to the majority of IT people, but some kind of application
> specific language tailored to the problem could be made palatable I suspect and
> also be efficient.

The whole methodology of working with Forth is based on developing an 
application-oriented language for the domain in question.

> However most Forth professionals are busy solving a completely different set of
> problems, so perhaps there is just not the critical mass of Forth people to
> to get activity in these other areas. (?)

It's not so much a problem of finding a "critical mass of Forth people" 
to work on a particular application domain, it's a matter of finding a 
funding source for it. Programmers, like other people, expect to eat 
dinner every day. Chuck's early work was funded by NRAO. He worked on 
the applications they paid him to do. When he and I founded FORTH, Inc., 
we made a living writing the applications our customers wanted.

Cheers,
Elizabeth

-- 
Elizabeth D. Rather
FORTH, Inc.
6080 Center Drive, Suite 600
Los Angeles, CA  90045
USA

[toc] | [prev] | [next] | [standalone]

#58078

In article <mJOdnfVoXsSSZ2PFnZ2dnUU7-ffNnZ2d@supernews.com>, 
erather@forth.com says...
> It's not so much a problem of finding a "critical mass of Forth people" 
> to work on a particular application domain, it's a matter of finding a 
> funding source for it. Programmers, like other people, expect to eat 
> dinner every day. Chuck's early work was funded by NRAO. He worked on 
> the applications they paid him to do. When he and I founded FORTH, Inc., 
> we made a living writing the applications our customers wanted.
> 
> Cheers,
> Elizabeth
> 
> 

A common mistake Elizabeth.

Take a look at free pascal,/lazarus or any number of other
projects. 

I'm doing my thing - it may come to nothing but there's
no third party paying me  (sad but that's life) 
 
All it takes is a few competent people to put their minds to finishing
a project. Gravity sets in once something of worth is 
demonstrated. Until then keep your head down and get on with it.

There's enough people here that do nothing but gossip like 
12 year old girls every day.  (and probably a few robots - or robotic people)

But when anyone can say that anything they do is "Forth" all you
will ever get is chaos.

Forth doesn't need a cute little logo - it needs an enforced trademark.
And if it isn't compliant with the official standard it doesn't get to call
itself Forth.

--

john

=========================
http://johntech.co.uk

"Bleeding Edge Forum"
http://johntech.co.uk/forum/ 

=========================

[toc] | [prev] | [next] | [standalone]

#58084

In the 80's there was a lot of entusiasm in Forth community. 
Then a dark gray curtain was closed - all that was counted was business.
And finally Fourth Dimension consequently was let down. 

I do think this development is unique for Forth. In other communities
there are no contradictions between amateurs and professionals .
Of course, a lot of programmers want to spoil their hobby by making
it to a profession, but that doesn't damage the society in other ways
that the competition makes them greedy minded - which really must
be an obstacle for creativity.

As said, some computer scientist despise Forth for it's simplicity and 
invent new fascilities and error-possibilities all the time. Of cource every
kind of application can be made in Forth, but to prepare for that relatively
few programmers must develop the intermediate software that constantly
is developed by millions of people in the industry.

In time Forth-like ideas may become the industrial standard but it's not 
much anybody can do to influent - except by using Forth.

A mascot may have just as much influence as anybody though, even without 
being able to write a single program. Asmarketting people is well aware of.

[toc] | [prev] | [next] | [standalone]

#58114

On Tuesday, April 25, 2017 at 2:23:15 AM UTC-7, john wrote:
> In article <mJOdnfVoXsSSZ2PFnZ2dnUU7-ffNnZ2d@supernews.com>, 
> erather@forth.com says...
> > It's not so much a problem of finding a "critical mass of Forth people" 
> > to work on a particular application domain, it's a matter of finding a 
> > funding source for it. Programmers, like other people, expect to eat 
> > dinner every day. Chuck's early work was funded by NRAO. He worked on 
> > the applications they paid him to do. When he and I founded FORTH, Inc., 
> > we made a living writing the applications our customers wanted.
> > 
> > Cheers,
> > Elizabeth
> 
> A common mistake Elizabeth.
> 
> Take a look at free pascal,/lazarus or any number of other
> projects. 
> 
> I'm doing my thing - it may come to nothing but there's
> no third party paying me  (sad but that's life) 
>  
> All it takes is a few competent people to put their minds to finishing
> a project. Gravity sets in once something of worth is 
> demonstrated. Until then keep your head down and get on with it.

I got paid to write MFX --- there was a lot of work involved --- I wouldn't have written that for free (and I'm not enthusiastic about writing Straight Forth for free now).

> But when anyone can say that anything they do is "Forth" all you
> will ever get is chaos.

MFX was not ANS-Forth compliant --- and it was only loosely compliant with Forth-83 (I had not, at that time, actually read the Forth-83 document) --- does this mean that I don't get to call it a "Forth" cross-compiler?

MFX did not result in chaos though --- MFX continues to be used today on the RACE processor, which is used in all of the motion-control boards sold by Testra --- how many motion-control boards has Forth Inc. built and sold? The only thing that Forth Inc. sells is hot air and blue sky, which seems to me to be closer to chaos than a functioning motion-control board.

> Forth doesn't need a cute little logo - it needs an enforced trademark.
> And if it isn't compliant with the official standard it doesn't get to call
> itself Forth.

How will you enforce this? If I say that MFX is "Forth" you will apparently hit me with a big lawsuit (similar to how Elizabeth Rather tried to hit Apple Computer with a big lawsuit because they were using the name "Swift" for their language) --- or, more likely, you will just flame me on the internet.

I say that my novice-package is Forth, and that it is fully ANS-Forth compliant, but all of Elizabeth Rather's sycophants say that it contradicts the "Forth Way" and hence I don't get to call it Forth. It seems obvious to me that if you allow the Forth Inc. marketing department (Elizabeth Rather) to decide what is "Forth" and what is not, then only Forth Inc. products will be "Forth" and everything else will be a non-standard wanna-bee.

Alex McDonald has spent almost 4 years now saying that I have a "serious misunderstanding of how pointers work" because my array sort and my array definer allows arrays of structs --- I haven't yet rewritten my array sort and my array definer to be restricted to only supporting arrays of cells --- so apparently the endless flaming doesn't work as well as hoped.

[toc] | [prev] | [next] | [standalone]

#58183

On 4/25/2017 19:25, hughaguilar96@gmail.com wrote:

> (and I'm not enthusiastic about
> writing Straight Forth for free now).

Start paying yourself, make yourself indispensable and then -- kaboom -- 
ask for a raise.

> 
> Alex McDonald has spent almost 4 years now saying that I have a
> "serious misunderstanding of how pointers work" because my array sort
> and my array definer allows arrays of structs --- *I haven't yet
> rewritten my array sort and my array definer to be restricted to only
> supporting arrays of cells* --- so apparently the endless flaming
> doesn't work as well as hoped.
> 

There you go again. Seriously misunderstanding.

-- 
Alex

[toc] | [prev] | [next] | [standalone]

#58186

On 26/04/2017 17:57, Alex wrote:
> On 4/25/2017 19:25, hughaguilar96@gmail.com wrote:
> 
>> (and I'm not enthusiastic about
>> writing Straight Forth for free now).
> 
> Start paying yourself, make yourself indispensable and then -- kaboom --
> ask for a raise.

Hey!  That's my business model!


>> Alex McDonald has spent almost 4 years now saying that I have a
>> "serious misunderstanding of how pointers work" because my array sort
>> and my array definer allows arrays of structs --- *I haven't yet
>> rewritten my array sort and my array definer to be restricted to only
>> supporting arrays of cells* --- so apparently the endless flaming
>> doesn't work as well as hoped.
>>
> 
> There you go again. Seriously misunderstanding.

Not so!  He understands something other than what you've said, that's all.

[toc] | [prev] | [next] | [standalone]

#58189

On 4/26/2017 16:41, Ron Aaron wrote:
> 
> 
> On 26/04/2017 17:57, Alex wrote:

>>>
>>
>> There you go again. Seriously misunderstanding.
> 
> Not so!  He understands something other than what you've said, that's all.
> 

I just had a Marx party of the third party moment there.

-- 
Alex

[toc] | [prev] | [next] | [standalone]

#58192

On 26/04/2017 19:59, Alex wrote:
> On 4/26/2017 16:41, Ron Aaron wrote:
>>
>>
>> On 26/04/2017 17:57, Alex wrote:
> 
>>>>
>>>
>>> There you go again. Seriously misunderstanding.
>>
>> Not so!  He understands something other than what you've said, that's
>> all.
>>
> 
> I just had a Marx party of the third party moment there.
> 

With a sanity clause?

[toc] | [prev] | [next] | [standalone]

#58213

On 4/26/2017 2:38 PM, Ron Aaron wrote:
>
>
> On 26/04/2017 19:59, Alex wrote:
>> On 4/26/2017 16:41, Ron Aaron wrote:
>>>
>>>
>>> On 26/04/2017 17:57, Alex wrote:
>>
>>>>>
>>>>
>>>> There you go again. Seriously misunderstanding.
>>>
>>> Not so!  He understands something other than what you've said, that's
>>> all.
>>>
>>
>> I just had a Marx party of the third party moment there.
>>
>
> With a sanity clause?

Everybody knows there is no sanity clause.  (an omage to the Marx Bros)

-- 

Rick C

[toc] | [prev] | [next] | [standalone]

Page 1 of 3  [1] 2 3  Next page →

Back to top | Article view | comp.lang.forth

csiph-web