1. Open source Euphoria...
Although I'm a fan of the open source philosophy, being at the "sharp end", so
to speak over the last few months has highlighted some issues which have taken
the shine off somewhat.
I speak from the point of view of a "Joe User", and it seems to me that the
focus has moved from that of the user to that of the developer. That's fine, but
I'd like to see (and I think this may have been mentioned before) a forum which
makes a distinction between the two. It can be intimidating for a newbie or
casual user to come to a forum where the content is dominated by detailed
technical discussion on the merits or otherwise of some change to Euphoria
internals, and reinforces the widely held notion that open source is for the
technocratic elite.
Secondly, I feel a certain disempowerment since Eu went open source, a certain
lack of control - which comes, paradoxically, from the perception that I have
more control (after all, Eu is open source, so it belongs to me as much as anyone
else right?). When it was proprietary product, I just payed my registration every
upgrade and forgot about it. It was Rob's business to take care of the bug fixes,
new features etc, and I trusted him to do that.
In a recent post (sequence ops) Rob finished with the words: "So what should we
do? That's up to you!" Well, that's great, but what if you don't have a extensive
knowledge of C, or a background in compiler design? and don't have the time
and/or inclination to learn?
Like most "users", all I'm really interested in is having a product which just
works, is reliable and (importantly) doesn't radically depart from the principles
which attracted me to it in the first place. That last point is worth
emphasising, I spent a considerable amount of time looking for a programming
language that suited my needs, and I found it in Euphoria. The last thing I want
to see is "feature creep".
One thing I've never really understood about programmers is this need to
ceaselessly "improve" their programs. Why is a program never just finished? There
comes a point when further "improvement" becomes feature creep. What's wrong with
having a clean and simple language which does a few things very well? This
concept is key to the longevity and success of Unix - simple tools which do one
thing well.
Sorry, this seems to have turned into a rant. :) but I'm sure I've articulated
some thoughts that are in the minds of other Eu users.
2. Re: Open source Euphoria...
My thoughts exactly. Haven't seem many newbie comments lately either. I do
however read these "technocratic" comments with interest though some are
scary. Actually I'm totally happy with 2.3 and the wonderful tools that
others have added to the library that extend EU's usefulness.
3. Re: Open source Euphoria...
Jules wrote:
>
> Although I'm a fan of the open source philosophy, being at the "sharp end",
> so to speak over the last few months has highlighted some issues which have
> taken the shine off somewhat.
>
> I speak from the point of view of a "Joe User", and it seems to me that the
> focus has moved from that of the user to that of the developer. That's fine,
> but I'd like to see (and I think this may have been mentioned before) a forum
> which makes a distinction between the two. It can be intimidating for a newbie
> or casual user to come to a forum where the content is dominated by detailed
> technical discussion on the merits or otherwise of some change to Euphoria
> internals,
> and reinforces the widely held notion that open source is for the technocratic
> elite.
>
> Secondly, I feel a certain disempowerment since Eu went open source, a certain
> lack of control - which comes, paradoxically, from the perception that I have
> more control (after all, Eu is open source, so it belongs to me as much as
> anyone
> else right?). When it was proprietary product, I just payed my registration
> every upgrade and forgot about it. It was Rob's business to take care of the
> bug fixes, new features etc, and I trusted him to do that.
>
> In a recent post (sequence ops) Rob finished with the words: "So what should
> we do? That's up to you!" Well, that's great, but what if you don't have a
> extensive
> knowledge of C, or a background in compiler design? and don't have the time
> and/or inclination to learn?
>
> Like most "users", all I'm really interested in is having a product which just
> works, is reliable and (importantly) doesn't radically depart from the
> principles
> which attracted me to it in the first place. That last point is worth
> emphasising,
> I spent a considerable amount of time looking for a programming language that
> suited my needs, and I found it in Euphoria. The last thing I want to see is
> "feature creep".
>
> One thing I've never really understood about programmers is this need to
> ceaselessly
> "improve" their programs. Why is a program never just finished? There comes
> a point when further "improvement" becomes feature creep. What's wrong with
> having a clean and simple language which does a few things very well? This
> concept
> is key to the longevity and success of Unix - simple tools which do one thing
> well.
>
> Sorry, this seems to have turned into a rant. :) but I'm sure I've articulated
> some thoughts that are in the minds of other Eu users.
Hi
Excellently put. Wholeheartedly agree.
And as I've said before, this is a list, not a forum, which would be ideally
suited to separating into different categories.
Chris
4. Re: Open source Euphoria...
Bernie
My files in archive:
WMOTOR, XMOTOR, W32ENGIN, MIXEDLIB, EU_ENGIN, WIN32ERU, WIN32API
Can be downloaded here:
http://www.rapideuphoria.com/cgi-bin/asearch.exu?dos=on&win=on&lnx=on&gen=on&keywords=bernie+ryan
5. Re: Open source Euphoria...
Despite my deep appreciation for many open source projects I
always liked Euphoria with Rob as a benevolent monarch.
A few weeks ago, in a private message, I remarked to Rob
how successful the transition to OpenEuphoria had been.
Almost immediately bug reports began to surface. Now
I have learned that bugs are a part of program development,
but what I now realize that Rob is no longer
going to be responsible for fixing these bugs.
Yesterday, I followed the link provided by Juergen to the
Euphoria bug list on SourceForge. It was small, but
immediately I got a very queasy feeling. Several of the
bugs listed where submitted by "nobody" and assigned to
"nobody".
I don't think we can have Falstaff and have him thin.
I'm no expert on Open Source software development, but
I am of the opinion based upon what I have observed that
it can be quite a chaotic process, fraught with its own
particular perils. It was not so long ago that the
development of new Linux features was put on hold because
the number of outstanding bugs had grown so large.
I think its time to have a Forum or a separate development
mailing list. Not everyone can stand the heat of the
open source development kitchen.
I really don't want to see a fly in my soup though! 
Ken Rhodes
Folding at Home: http://folding.stanford.edu/
100% MicroSoft Free
SuSE Linux 10.0
No AdWare, SpyWare, or Viruses!
Life is Good, 
6. Re: Open source Euphoria...
Jules wrote:
>
> Although I'm a fan of the open source philosophy, being at the "sharp end",
> so to speak over the last few months has highlighted some issues which have
> taken the shine off somewhat.
>
I wholeheartedly disagree with your analysis, but let us see why in more detail.
> I speak from the point of view of a "Joe User", and it seems to me that the
> focus has moved from that of the user to that of the developer. That's fine,
> but I'd like to see (and I think this may have been mentioned before) a forum
> which makes a distinction between the two. It can be intimidating for a newbie
> or casual user to come to a forum where the content is dominated by detailed
> technical discussion on the merits or otherwise of some change to Euphoria
> internals,
> and reinforces the widely held notion that open source is for the technocratic
> elite.
I'll second o you on this one. The only potential pitfall is that developers
might forget to go read the user forum often enough. We must be aware of it and
proceed anyway. This has been a recurring demand.
>
> Secondly, I feel a certain disempowerment since Eu went open source, a certain
> lack of control - which comes, paradoxically, from the perception that I have
> more control (after all, Eu is open source, so it belongs to me as much as
> anyone
> else right?).
Do you really need being controlled? I feel sorry for you. Or misunderstood you.
> When it was proprietary product, I just payed my registration
> every upgrade and forgot about it. It was Rob's business to take care of the
> bug fixes, new features etc, and I trusted him to do that.
And oft requested features weren't introduced.
>
> In a recent post (sequence ops) Rob finished with the words: "So what should
> we do? That's up to you!" Well, that's great, but what if you don't have a
> extensive
> knowledge of C, or a background in compiler design? and don't have the time
> and/or inclination to learn?
The whole community doesn't need to have the time, skill and knowledge. Rob said
it better than I in a rather recent post about cooperative problem solving.
I think our current problem is the lack of enough people with time, knowledge
and motivation. I've been here for only 5 years, and have seen a fair amount of
people with these abilities leaving because the language had been stagnating so
much.
Further, did Rob hint that he would no longer help in the continuing
development? I don't think he did.
>
> Like most "users", all I'm really interested in is having a product which just
> works, is reliable and (importantly) doesn't radically depart from the
> principles
> which attracted me to it in the first place. That last point is worth
> emphasising,
> I spent a considerable amount of time looking for a programming language that
> suited my needs, and I found it in Euphoria. The last thing I want to see is
> "feature creep".
>
I'd describe the current state as "feature catch up" rather. Feature creep isn't
good, but is still very remote.
Things that wouldn't benefit much from being done inside the interpreter shoud
be left outside so as to be used on demand, and without any impact on the
interpreter, by including files, be them standard or not. But abilities that
cannot be implemented efficiently enough from outside should be implemented
inside instead of being simply left out.
> One thing I've never really understood about programmers is this need to
> ceaselessly
> "improve" their programs. Why is a program never just finished? There comes
> a point when further "improvement" becomes feature creep. What's wrong with
> having a clean and simple language which does a few things very well?
The fact that it does so many things so awkwardly, or not at all, besides.
Programs which hardly anyone uses get finished and gather dust. Programs that
are useful are used up to and beyond their initial scope, limits and intent,
stirring demands for pushing them further. So, as you are asking "why", I'd
answer: because their products and themselves are alive, not lying dead on a
shelf. Growth is healthy, overgrowth eventually kills, undergrowth surelykills.
> This concept
> is key to the longevity and success of Unix - simple tools which do one thing
> well.
You said success? In some very specific software compartments, mostly networking
as far as I know, that's true. In the public at large, I don't think the word
faithfully describes the reality.
>
> Sorry, this seems to have turned into a rant. :) but I'm sure I've articulated
> some thoughts that are in the minds of other Eu users.
Rants hardly help, confrontation and discussion often do, in my experience.
CChris
7. Re: Open source Euphoria...
CChris wrote:
>
> Jules wrote:
> >
> > Although I'm a fan of the open source philosophy, being at the "sharp end",
> > so to speak over the last few months has highlighted some issues which have
> > taken the shine off somewhat.
> >
>
> I wholeheartedly disagree with your analysis, but let us see why in more
> detail.
>
> > I speak from the point of view of a "Joe User", and it seems to me that the
> > focus has moved from that of the user to that of the developer. That's fine,
> > but I'd like to see (and I think this may have been mentioned before) a
> > forum
> > which makes a distinction between the two. It can be intimidating for a
> > newbie
> > or casual user to come to a forum where the content is dominated by detailed
> > technical discussion on the merits or otherwise of some change to Euphoria
> > internals,
> > and reinforces the widely held notion that open source is for the
> > technocratic
> > elite.
>
> I'll second o you on this one. The only potential pitfall is that developers
> might forget to go read the user forum often enough. We must be aware of it
> and proceed anyway. This has been a recurring demand.
>
> >
> > Secondly, I feel a certain disempowerment since Eu went open source, a
> > certain
> > lack of control - which comes, paradoxically, from the perception that I
> > have
> > more control (after all, Eu is open source, so it belongs to me as much as
> > anyone
> > else right?).
>
> Do you really need being controlled? I feel sorry for you. Or misunderstood
> you.
I probably could have been more clear on this point. Maybe an analogy will help.
If I was a mechanic, why would I take my car to a garage for maintainence,
repairs, or adding a rear spoiler? But I'm not a mechanic, nor do I particular
enjoy digging around under the hood. I just use my car as a means of getting from
A to B. I'd rather pay my local mechanic so I don't have to think about anything
other than that.
I think "controlled" is one of those value-laden words. We are all "controlled"
in one way or another. I prefer to say "I would rather let someone else take that
responsibility so I don't have to think about it".
> > When it was proprietary product, I just payed my registration
> > every upgrade and forgot about it. It was Rob's business to take care of the
> > bug fixes, new features etc, and I trusted him to do that.
>
> And oft requested features weren't introduced.
I never had a problem with that, maybe because I'm not a professional programmer
and haven't had experiences with other programming languages (Only BASIC prior to
Eu), so didn't really know what I was missing, perhaps.
>
> >
> > In a recent post (sequence ops) Rob finished with the words: "So what should
> > we do? That's up to you!" Well, that's great, but what if you don't have a
> > extensive
> > knowledge of C, or a background in compiler design? and don't have the time
> > and/or inclination to learn?
>
> The whole community doesn't need to have the time, skill and knowledge. Rob
> said it better than I in a rather recent post about cooperative problem
> solving.
I'd like to read that post.
>
> I think our current problem is the lack of enough people with time, knowledge
> and motivation. I've been here for only 5 years, and have seen a fair amount
> of people with these abilities leaving because the language had been
> stagnating
> so much.
>
> Further, did Rob hint that he would no longer help in the continuing
> development?
> I don't think he did.
>
> >
> > Like most "users", all I'm really interested in is having a product which
> > just
> > works, is reliable and (importantly) doesn't radically depart from the
> > principles
> > which attracted me to it in the first place. That last point is worth
> > emphasising,
> > I spent a considerable amount of time looking for a programming language
> > that
> > suited my needs, and I found it in Euphoria. The last thing I want to see is
> > "feature creep".
> >
>
> I'd describe the current state as "feature catch up" rather. Feature creep
> isn't
> good, but is still very remote.
>
> Things that wouldn't benefit much from being done inside the interpreter shoud
> be left outside so as to be used on demand, and without any impact on the
> interpreter,
> by including files, be them standard or not. But abilities that cannot be
> implemented
> efficiently enough from outside should be implemented inside instead of being
> simply left out.
But everyone has different ideas about what features should or should not be
implemented - and relatively few of us are able to understand the consequences of
this or that feature. I like the minimalist nature of Euphoria, sometimes I'll go
weeks without coding - it's very easy to pick up after an absence.
>
> > One thing I've never really understood about programmers is this need to
> > ceaselessly
> > "improve" their programs. Why is a program never just finished? There comes
> > a point when further "improvement" becomes feature creep. What's wrong with
> > having a clean and simple language which does a few things very well?
>
> The fact that it does so many things so awkwardly, or not at all, besides.
>
> Programs which hardly anyone uses get finished and gather dust. Programs that
> are useful are used up to and beyond their initial scope, limits and intent,
> stirring demands for pushing them further. So, as you are asking "why", I'd
> answer: because their products and themselves are alive, not lying dead on a
> shelf. Growth is healthy, overgrowth eventually kills, undergrowth
> surelykills.
I guess I just don't get it. If something is used and found to be useful is it
"dead" because it isn't being continually improved?
>
> > This concept
> > is key to the longevity and success of Unix - simple tools which do one
> > thing
> > well.
>
> You said success? In some very specific software compartments, mostly
> networking
> as far as I know, that's true. In the public at large, I don't think the word
> faithfully describes the reality.
>
> >
> > Sorry, this seems to have turned into a rant. :) but I'm sure I've
> > articulated
> > some thoughts that are in the minds of other Eu users.
>
> Rants hardly help, confrontation and discussion often do, in my experience.
>
well, we are having a discussion aren't we? :)
If Rob hasn't made changes under pressure before Eu went open source, why would
he do so now? he hasn't come up with any suggestions himself regarding
improvements, but rather reacts only to suggestions made by others (this is not a
criticism btw).
8. Re: Open source Euphoria...
Kenneth Rhodes wrote:
>
> Despite my deep appreciation for many open source projects I
> always liked Euphoria with Rob as a benevolent monarch.
Me too.
> Yesterday, I followed the link provided by Juergen to the
> Euphoria bug list on SourceForge. It was small, but
> immediately I got a very queasy feeling. Several of the
> bugs listed where submitted by "nobody" and assigned to
> "nobody".
>
Agreed, I had the very same feeling. I thought one of the benefits of open
source was that bugs were fixed more quickly, but of course, that only works when
there are enough people capable of fixing them.
9. Re: Open source Euphoria...
Kenneth Rhodes wrote:
> Despite my deep appreciation for many open source projects I
> always liked Euphoria with Rob as a benevolent monarch.
>
> A few weeks ago, in a private message, I remarked to Rob
> how successful the transition to OpenEuphoria had been.
> Almost immediately bug reports began to surface. Now
> I have learned that bugs are a part of program development,
> but what I now realize that Rob is no longer
> going to be responsible for fixing these bugs.
>
> Yesterday, I followed the link provided by Juergen to the
> Euphoria bug list on SourceForge. It was small, but
> immediately I got a very queasy feeling. Several of the
> bugs listed where submitted by "nobody"
One "submission by nobody" was due to me. Here's the explanation:
<http://www.openeuphoria.org/cgi-bin/esearch.exu?fromMonth=6&fromYear=C&toMonth=6&toYear=C&postedBy=Matt+Lewis&keywords=17%3A36>
> and assigned to "nobody".
What does that actually mean: a bug is "assigned to someone"?
> I don't think we can have Falstaff and have him thin.
> I'm no expert on Open Source software development, but
> I am of the opinion based upon what I have observed that
> it can be quite a chaotic process, fraught with its own
> particular perils. It was not so long ago that the
> development of new Linux features was put on hold because
> the number of outstanding bugs had grown so large.
>
> I think its time to have a Forum or a separate development
> mailing list. Not everyone can stand the heat of the
> open source development kitchen.
>
> I really don't want to see a fly in my soup though!
I think we need as soon as possible some kind of organization,
some plan or a "set of rules" or something. Open source developement
does not work automatically on its own.
Regards,
Juergen
10. Re: Open source Euphoria...
CChris wrote:
>
> Jules wrote:
<SNIP>
> > Secondly, I feel a certain disempowerment since Eu went open source, a
> > certain
> > lack of control - which comes, paradoxically, from the perception that I
> > have
> > more control (after all, Eu is open source, so it belongs to me as much as
> > anyone
> > else right?).
>
> Do you really need being controlled? I feel sorry for you. Or misunderstood
> you.
<SNIP>
> >
Yes, you misunderstood, I think. Jules was expressing something hard to
get across, a feeling or intuition. Saying it a different way, and speaking
for myself, I would say it is the sense of possible loss of integrity in the
process of Eu development (not moral integrity, just the integrity of a
single person, as compared with a group).
> > Sorry, this seems to have turned into a rant. :) but I'm sure I've
> > articulated
> > some thoughts that are in the minds of other Eu users.
>
> Rants hardly help, confrontation and discussion often do, in my experience.
>
> CChris
Right. Hence my throwing my opinion in the mix. I do appreciate the
comments of Jules and Kenneth Rhodes. However I admit I have never been
able to come down on one side or the other of the "forking" of the Eu
forum according to some criteria -- level of expertise or interest or
whatever. Many advantages and disadvantages either way, but even though
many do not post, they always can and they should. CChris expressed concern
those of more expertise might not read the comments of those with less, but
the reverse is a possibility too.
As to the integrity of the development of Euphoria, I wonder if it has to
do with something Derek has stated and repeated:
"The case for breaking existing code must be based on actual empirical usage
and not guessitmates on what and who will be affected.
"If it turns out that the cost of fixing existing code is too steep then we
won't change, but if it is not more costly than perpetually having
non-intuitive constructs then change should be considered."
If changes in Eu code were done based on empirical evidence in favor of
a proposed improvement, and demonstrated to the community as a better way,
then the integrity problem would go away, and those of lesser expertise
(like me) would feel more a part of the process.
--Quark
11. Re: Open source Euphoria...
Jules wrote:
>
> Kenneth Rhodes wrote:
> >
> > Despite my deep appreciation for many open source projects I
> > always liked Euphoria with Rob as a benevolent monarch.
>
> Me too.
>
> > Yesterday, I followed the link provided by Juergen to the
> > Euphoria bug list on SourceForge. It was small, but
> > immediately I got a very queasy feeling. Several of the
> > bugs listed where submitted by "nobody" and assigned to
> > "nobody".
> >
> Agreed, I had the very same feeling. I thought one of the benefits of open
> source
> was that bugs were fixed more quickly, but of course, that only works when
> there
> are enough people capable of fixing them.
ME TOO !
Bernie
My files in archive:
WMOTOR, XMOTOR, W32ENGIN, MIXEDLIB, EU_ENGIN, WIN32ERU, WIN32API
Can be downloaded here:
http://www.rapideuphoria.com/cgi-bin/asearch.exu?dos=on&win=on&lnx=on&gen=on&keywords=bernie+ryan
12. Re: Open source Euphoria...
DB James wrote:
>
> CChris wrote:
> >
> > Jules wrote:
> <SNIP>
> > > Secondly, I feel a certain disempowerment since Eu went open source, a
> > > certain
> > > lack of control - which comes, paradoxically, from the perception that I
> > > have
> > > more control (after all, Eu is open source, so it belongs to me as much as
> > > anyone
> > > else right?).
> >
> > Do you really need being controlled? I feel sorry for you. Or misunderstood
> > you.
> <SNIP>
> > >
>
> Yes, you misunderstood, I think. Jules was expressing something hard to
> get across, a feeling or intuition. Saying it a different way, and speaking
> for myself, I would say it is the sense of possible loss of integrity in the
> process of Eu development (not moral integrity, just the integrity of a
> single person, as compared with a group).
yes, It was hard to get across, but you've done it. :)
>
> > > Sorry, this seems to have turned into a rant. :) but I'm sure I've
> > > articulated
> > > some thoughts that are in the minds of other Eu users.
> >
> > Rants hardly help, confrontation and discussion often do, in my experience.
> >
> > CChris
>
> Right. Hence my throwing my opinion in the mix. I do appreciate the
> comments of Jules and Kenneth Rhodes. However I admit I have never been
> able to come down on one side or the other of the "forking" of the Eu
> forum according to some criteria -- level of expertise or interest or
> whatever. Many advantages and disadvantages either way, but even though
> many do not post, they always can and they should. CChris expressed concern
> those of more expertise might not read the comments of those with less, but
> the reverse is a possibility too.
>
> As to the integrity of the development of Euphoria, I wonder if it has to
> do with something Derek has stated and repeated:
>
> "The case for breaking existing code must be based on actual empirical usage
> and not guessitmates on what and who will be affected.
>
> "If it turns out that the cost of fixing existing code is too steep then we
> won't change, but if it is not more costly than perpetually having
> non-intuitive constructs then change should be considered."
>
> If changes in Eu code were done based on empirical evidence...
Not sure what you mean by "empirical evidence"
13. Re: Open source Euphoria...
Bernie Ryan wrote:
> Jules wrote:
> > Agreed, I had the very same feeling. I thought one of the benefits of open
> > source
> > was that bugs were fixed more quickly, but of course, that only works when
> > there
> > are enough people capable of fixing them.
> ME TOO !
Rob, you really let us down letting Euphoria go open source.
14. Re: Open source Euphoria...
Juergen Luethje wrote:
>
> Kenneth Rhodes wrote:
>
> > Despite my deep appreciation for many open source projects I
> > always liked Euphoria with Rob as a benevolent monarch.
> >
> > A few weeks ago, in a private message, I remarked to Rob
> > how successful the transition to OpenEuphoria had been.
> > Almost immediately bug reports began to surface. Now
> > I have learned that bugs are a part of program development,
> > but what I now realize that Rob is no longer
> > going to be responsible for fixing these bugs.
> >
> > Yesterday, I followed the link provided by Juergen to the
> > Euphoria bug list on SourceForge. It was small, but
> > immediately I got a very queasy feeling. Several of the
> > bugs listed where submitted by "nobody"
>
> One "submission by nobody" was due to me. Here's the explanation:
> <<a
> href="http://www.openeuphoria.org/cgi-bin/esearch.exu?fromMonth=6&fromYear=C&toMonth=6&toYear=C&postedBy=Matt+Lewis&keywords=17%3A36">http://www.openeuphoria.org/cgi-bin/esearch.exu?fromMonth=6&fromYear=C&toMonth=6&toYear=C&postedBy=Matt+Lewis&keywords=17%3A36</a>>
>
ok - now I remember reading that post!
> > and assigned to "nobody".
>
> What does that actually mean: a bug is "assigned to someone"?
Looks that way according to the chart:
http://sourceforge.net/tracker/?group_id=182827&atid=902782
One bug is assigned to Matt Lewis. The other 4 listed do not appear
to be assigned to anyone.
Perhaps there should be a default assignment to a member of the
development team designated as the "bug list supervisor" who can
then assign task of fixing the bug to one of the developers. I
believe that Rob mentioned in a post to the EuForum that the
person who implemented the bug should be responsible for fixing
it - or something to that effect.
My impression is that Rob is trying not to be controlling
in the transition of Euphoria to to an opensource development
model. I think he is going to have to devote a pretty good bit
of time to being an administrator on the front end though.
>
> > I don't think we can have Falstaff and have him thin.
> > I'm no expert on Open Source software development, but
> > I am of the opinion based upon what I have observed that
> > it can be quite a chaotic process, fraught with its own
> > particular perils. It was not so long ago that the
> > development of new Linux features was put on hold because
> > the number of outstanding bugs had grown so large.
> >
> > I think its time to have a Forum or a separate development
> > mailing list. Not everyone can stand the heat of the
> > open source development kitchen.
> >
> > I really don't want to see a fly in my soup though!
>
> I think we need as soon as possible some kind of organization,
> some plan or a "set of rules" or something. Open source developement
> does not work automatically on its own.
>
I agree.
Ken Rhodes
Folding at Home: http://folding.stanford.edu/
100% MicroSoft Free
SuSE Linux 10.0
No AdWare, SpyWare, or Viruses!
Life is Good,
15. Re: Open source Euphoria...
Jules wrote:
>
> CChris wrote:
> >
> > Jules wrote:
> > >
> > > Although I'm a fan of the open source philosophy, being at the "sharp
> > > end",
> > > so to speak over the last few months has highlighted some issues which
> > > have
> > > taken the shine off somewhat.
> > >
> >
> > I wholeheartedly disagree with your analysis, but let us see why in more
> > detail.
> >
> > > I speak from the point of view of a "Joe User", and it seems to me that
> > > the
> > > focus has moved from that of the user to that of the developer. That's
> > > fine,
> > > but I'd like to see (and I think this may have been mentioned before) a
> > > forum
> > > which makes a distinction between the two. It can be intimidating for a
> > > newbie
> > > or casual user to come to a forum where the content is dominated by
> > > detailed
> > > technical discussion on the merits or otherwise of some change to Euphoria
> > > internals,
> > > and reinforces the widely held notion that open source is for the
> > > technocratic
> > > elite.
> >
> > I'll second o you on this one. The only potential pitfall is that developers
> > might forget to go read the user forum often enough. We must be aware of it
> > and proceed anyway. This has been a recurring demand.
> >
> > >
> > > Secondly, I feel a certain disempowerment since Eu went open source, a
> > > certain
> > > lack of control - which comes, paradoxically, from the perception that I
> > > have
> > > more control (after all, Eu is open source, so it belongs to me as much as
> > > anyone
> > > else right?).
> >
> > Do you really need being controlled? I feel sorry for you. Or misunderstood
> > you.
>
> I probably could have been more clear on this point. Maybe an analogy will
> help.
> If I was a mechanic, why would I take my car to a garage for maintainence,
> repairs,
> or adding a rear spoiler? But I'm not a mechanic, nor do I particular enjoy
> digging around under the hood. I just use my car as a means of getting from
> A to B. I'd rather pay my local mechanic so I don't have to think about
> anything
> other than that.
> I think "controlled" is one of those value-laden words. We are all
> "controlled"
> in one way or another. I prefer to say "I would rather let someone else take
> that responsibility so I don't have to think about it".
>
Ok. Now you have more mechanics to talk to. And not getting involved in
something you use or need is imho the best definition of disempowerment, so we
agree, at least formally.
> > > When it was proprietary product, I just payed my registration
> > > every upgrade and forgot about it. It was Rob's business to take care of
> > > the
> > > bug fixes, new features etc, and I trusted him to do that.
> >
> > And oft requested features weren't introduced.
>
> I never had a problem with that, maybe because I'm not a professional
> programmer
> and haven't had experiences with other programming languages (Only BASIC prior
> to Eu), so didn't really know what I was missing, perhaps.
I'm not a professional either.
What may make me think differently from a large part of the community is that I
was never exposed to BASIC. Oh yes, in high school, 25 years ago.
> >
> > >
> > > In a recent post (sequence ops) Rob finished with the words: "So what
> > > should
> > > we do? That's up to you!" Well, that's great, but what if you don't have a
> > > extensive
> > > knowledge of C, or a background in compiler design? and don't have the
> > > time
> > > and/or inclination to learn?
> >
> > The whole community doesn't need to have the time, skill and knowledge. Rob
> > said it better than I in a rather recent post about cooperative problem
> > solving.
>
> I'd like to read that post.
Date: 2007 Jul 2 17:12
Unfortunately, the exact message number doesn't appear in the search EuForum
pages, or I didn't see it. (Ping Rob...)
> >
> > I think our current problem is the lack of enough people with time,
> > knowledge
> > and motivation. I've been here for only 5 years, and have seen a fair amount
> > of people with these abilities leaving because the language had been
> > stagnating
> > so much.
> >
> > Further, did Rob hint that he would no longer help in the continuing
> > development?
> > I don't think he did.
> >
> > >
> > > Like most "users", all I'm really interested in is having a product which
> > > just
> > > works, is reliable and (importantly) doesn't radically depart from the
> > > principles
> > > which attracted me to it in the first place. That last point is worth
> > > emphasising,
> > > I spent a considerable amount of time looking for a programming language
> > > that
> > > suited my needs, and I found it in Euphoria. The last thing I want to see
> > > is
> > > "feature creep".
> > >
> >
> > I'd describe the current state as "feature catch up" rather. Feature creep
> > isn't
> > good, but is still very remote.
> >
> > Things that wouldn't benefit much from being done inside the interpreter
> > shoud
> > be left outside so as to be used on demand, and without any impact on the
> > interpreter,
> > by including files, be them standard or not. But abilities that cannot be
> > implemented
> > efficiently enough from outside should be implemented inside instead of
> > being
> > simply left out.
>
> But everyone has different ideas about what features should or should not be
> implemented - and relatively few of us are able to understand the consequences
> of this or that feature. I like the minimalist nature of Euphoria, sometimes
> I'll go weeks without coding - it's very easy to pick up after an absence.
Well, this very minimalism caused me to consider leaving it a few times, and it
may still happen.
I don't think introducing a feature often has major consequences on further use
and development. These fears are apparent, largely unsubstantiated and probably
harmful.
> >
> > > One thing I've never really understood about programmers is this need to
> > > ceaselessly
> > > "improve" their programs. Why is a program never just finished? There
> > > comes
> > > a point when further "improvement" becomes feature creep. What's wrong
> > > with
> > > having a clean and simple language which does a few things very well?
> >
> > The fact that it does so many things so awkwardly, or not at all, besides.
> >
> > Programs which hardly anyone uses get finished and gather dust. Programs
> > that
> > are useful are used up to and beyond their initial scope, limits and intent,
> > stirring demands for pushing them further. So, as you are asking "why", I'd
> > answer: because their products and themselves are alive, not lying dead on a
> > shelf. Growth is healthy, overgrowth eventually kills, undergrowth
> > surelykills.
>
> I guess I just don't get it. If something is used and found to be useful is
> it "dead" because it isn't being continually improved?
If it is useful, its range of uses will be extended by users. And that's how it
will improve.
> >
> > > This concept
> > > is key to the longevity and success of Unix - simple tools which do one
> > > thing
> > > well.
> >
> > You said success? In some very specific software compartments, mostly
> > networking
> > as far as I know, that's true. In the public at large, I don't think the
> > word
> > faithfully describes the reality.
> >
> > >
> > > Sorry, this seems to have turned into a rant. :) but I'm sure I've
> > > articulated
> > > some thoughts that are in the minds of other Eu users.
> >
> > Rants hardly help, confrontation and discussion often do, in my experience.
> >
> well, we are having a discussion aren't we? :)
>
Yes :)
> If Rob hasn't made changes under pressure before Eu went open source, why
> would
> he do so now? he hasn't come up with any suggestions himself regarding
> improvements,
> but rather reacts only to suggestions made by others (this is not a criticism
> btw).
He has positioned himself as benevolent overseer, and I wouldn't criticise this.
CChris
16. Re: Open source Euphoria...
I personally prefer the minimalism designed into the language and recent
discussions remind me of why the language behaves the way it does under certain
conditions and that it would be folly to change things without good reason.
That doesn't mean that nothing should change, however.
The move to open source has not decreased the quality of the language nor has it
caused previous versions to suddenly stop working.
It might be a good idea to have two separate discussions lists though (since we
don't seem to be moving to a forum anytime). One list for developers and people
who are interested in discussing possible changes to the language and one list
for actual programming help, tips, and announcements.
I'd still read both lists, as long as there was still an RSS feed.
--
"Any programming problem can be solved by adding a level of indirection."
--anonymous
"Any performance problem can be solved by removing a level of indirection."
--M. Haertel
"Premature optimization is the root of all evil in programming."
--C.A.R. Hoare
j.
17. Re: Open source Euphoria...
DB James wrote:
>
> CChris wrote:
> >
> > Jules wrote:
> <SNIP>
> > > Secondly, I feel a certain disempowerment since Eu went open source, a
> > > certain
> > > lack of control - which comes, paradoxically, from the perception that I
> > > have
> > > more control (after all, Eu is open source, so it belongs to me as much as
> > > anyone
> > > else right?).
> >
> > Do you really need being controlled? I feel sorry for you. Or misunderstood
> > you.
> <SNIP>
> > >
>
> Yes, you misunderstood, I think. Jules was expressing something hard to
> get across, a feeling or intuition. Saying it a different way, and speaking
> for myself, I would say it is the sense of possible loss of integrity in the
> process of Eu development (not moral integrity, just the integrity of a
> single person, as compared with a group).
>
Well... I tend to mistrust any single person by instinct, and found out that
groups err less often, unless they are too large. So small wonder I
misunderstood, since it seems that I don't have the same intuitions <smile/>
Sorry for that, Jules.
> > > Sorry, this seems to have turned into a rant. :) but I'm sure I've
> > > articulated
> > > some thoughts that are in the minds of other Eu users.
> >
> > Rants hardly help, confrontation and discussion often do, in my experience.
> >
> > CChris
>
> Right. Hence my throwing my opinion in the mix. I do appreciate the
> comments of Jules and Kenneth Rhodes. However I admit I have never been
> able to come down on one side or the other of the "forking" of the Eu
> forum according to some criteria -- level of expertise or interest or
> whatever. Many advantages and disadvantages either way, but even though
> many do not post, they always can and they should. CChris expressed concern
> those of more expertise might not read the comments of those with less, but
> the reverse is a possibility too.
>
Yes, but far less serious.
I have jnown Joe User for years, and his habit is not to care until he happens
to need something unusual. And be surprised that it doesn't come out of the box.
Since developers are supposed to work for the community, they ought to stay
informed of the users' requests, while the users only should keep an eye on
developments.
> As to the integrity of the development of Euphoria, I wonder if it has to
> do with something Derek has stated and repeated:
>
> "The case for breaking existing code must be based on actual empirical usage
> and not guessitmates on what and who will be affected.
>
> "If it turns out that the cost of fixing existing code is too steep then we
> won't change, but if it is not more costly than perpetually having
> non-intuitive constructs then change should be considered."
>
> If changes in Eu code were done based on empirical evidence in favor of
> a proposed improvement, and demonstrated to the community as a better way,
> then the integrity problem would go away, and those of lesser expertise
> (like me) would feel more a part of the process.
>
> --Quark
Definitely.
Problem is, if someone knows only about one sort of programming/apps, that
someone will usually reject whatever isn't of direct use to him. Hence, even
empirical evidence, which I have pushed for being the almost only criterium to
assess a change, may not be enough.
That's why I also usually promote additions, not changes that would break
existing code. I'm not against them, but think there should be a very solid
reason and large benefits to do so.
CChris
18. Re: Open source Euphoria...
- Posted by Matt Lewis <matthewwalkerlewis at ?mai?.com>
Jul 18, 2007
-
Last edited Jul 19, 2007
Kenneth Rhodes wrote:
>
> Juergen Luethje wrote:
> >
> ok - now I remember reading that post!
>
> > > and assigned to "nobody".
> >
> > What does that actually mean: a bug is "assigned to someone"?
It basically means that someone has agreed to work on the bug. Rob's
philosophy is that whoever created the bug should fix it. And it's
hard to disagree with that. I've been trying to find some time lately to
fix some of these things, but haven't managed to get around to it.
I did fix the file extension thing (nearly a one-liner--yay!). It was
really a case of an incorrect error message.
Matt
19. Re: Open source Euphoria...
- Posted by c.k.lester <euphoric at ck?ester?com>
Jul 18, 2007
-
Last edited Jul 19, 2007
Matt Lewis wrote:
>
>
> I did fix the file extension thing (nearly a one-liner--yay!). It was
> really a case of an incorrect error message.
When is anybody going to fix the error that won't let me hook up Euphoria to
my brain and let me program with thought?
Huh?!!?
20. Re: Open source Euphoria...
- Posted by Jason Gade <jaygade at ?ahoo.?om>
Jul 18, 2007
-
Last edited Jul 19, 2007
c.k.lester wrote:
>
> Matt Lewis wrote:
> >
> >
> > I did fix the file extension thing (nearly a one-liner--yay!). It was
> > really a case of an incorrect error message.
>
> When is anybody going to fix the error that won't let me hook up Euphoria to
> my brain and let me program with thought?
>
> Huh?!!?
'Cause it's not a software problem -- it's a hardware problem.
:D
--
"Any programming problem can be solved by adding a level of indirection."
--anonymous
"Any performance problem can be solved by removing a level of indirection."
--M. Haertel
"Premature optimization is the root of all evil in programming."
--C.A.R. Hoare
j.
21. Re: Open source Euphoria...
- Posted by Craig Welch <euphoriah at c?elc?.org>
Jul 18, 2007
-
Last edited Jul 19, 2007
Jules wrote:
> Like most "users", all I'm really interested in is having a product which just
> works, is reliable and (importantly) doesn't radically depart from the
> principles
> which attracted me to it in the first place. That last point is worth
> emphasising,
> I spent a considerable amount of time looking for a programming language that
> suited my needs, and I found it in Euphoria. The last thing I want to see is
> "feature creep".
I agree. When Euphoria went Open Source, I decided to stop 'upgrading' and stay
with 2.5. It works.
--
Craig
22. Re: Open source Euphoria...
- Posted by jacques deschênes <desja at glob?trotte?.net>
Jul 18, 2007
-
Last edited Jul 19, 2007
Rob as invested much much more in euphoria than everone else, personnaly I can
understand that I at some point he was tired and wanted to drift out of it.
Now that it is open source, if some programmers are really interested in the
future of euphoria its time for them to take it further.
If its not the case euphoria will simply die. Well! nothing last forever.
regards,
Jacques Deschênes
c.k.lester wrote:
>
> Bernie Ryan wrote:
> > Jules wrote:
> > > Agreed, I had the very same feeling. I thought one of the benefits of open
> > > source
> > > was that bugs were fixed more quickly, but of course, that only works when
> > > there
> > > are enough people capable of fixing them.
> > ME TOO !
>
> Rob, you really let us down letting Euphoria go open source.
23. Re: Open source Euphoria...
- Posted by c.k.lester <euphoric at c?lester?com>
Jul 18, 2007
-
Last edited Jul 19, 2007
jacques deschênes wrote:
> c.k.lester wrote:
> > Rob, you really let us down letting Euphoria go open source.
> Rob as invested much much more in euphoria than everone else, personnaly I can
> understand that I at some point he was tired and wanted to drift out of it.
> Now that it is open source, if some programmers are really interested in the
> future of euphoria its time for them to take it further.
> If its not the case euphoria will simply die. Well! nothing last forever.
I hope everyone knows I was trying to be ironic. With so much clamoring for
open source, to have someone be let down by that is hilarious. Right?!?! :P
24. Re: Open source Euphoria...
- Posted by DB James <larches at comca?t.?et>
Jul 18, 2007
-
Last edited Jul 19, 2007
Jules wrote:
>
> DB James wrote:
> >
> > CChris wrote:
> > >
> > > Jules wrote:
> > <SNIP>
> >
> > If changes in Eu code were done based on empirical evidence...
>
> Not sure what you mean by "empirical evidence"
Derek or CChris or others could answer this better, but an example (kindly
ignore errors) might be:
Ima Coder loves the "enum" keyword from C and she wants to have a keyword in
Eu called constantenum that works this way:
constantenum Name=1,Adr1,Adr2,Adr3,Phone,Last --imagine this 50 items long
sequence database,record
record={}
for i=1 to Last do record=append(record,{}) end for
database=repeat(record,1000)
database[1][Name]="Elvis Presley"
database[1][Adr1]="123 Blue Suede Lane"
...
Okay, so she implements this in a library such that when the code is
processed, it turns the following:
--constantenum Name=1,Adr1,Adr2,Adr3,Phone,Extra
to:
constant Name=1,Adr1=2,Adr2=3,Adr3=4,Phone=5,Extra=6
People try it and report variously: "great!", "dumb", ,"unnecessary",etc.
So, she gets serious, produces some code and puts it in a test version
of Euphoria. She demonstrates that her code would save time in program
development, etc. But some find it slows down the interpreter. She re-codes
and now demonstrates that it is fast. But some might show that it would
be better to simply borrow "enum" as used in C because (they may prove) it
would be more flexible. And on like that until there is a general consensus
to accept the code, change it, or forget it. (Or if the serious coders made
the decision in actual practice, fine.)
--Quark
25. Re: Open source Euphoria...
jacques deschênes wrote:
>
> Rob as invested much much more in euphoria than everone else, personnaly I can
> understand that I at some point he was tired and wanted to drift out of it.
<SNIP>
> > Rob, you really let us down letting Euphoria go open source.
Jacques, it's his own baby so he's not tired of it. But the kid is now old
enough to go to college, and he may figure it is time for it to stand on its
own.
ck, the whole thing goes back to a plot to assassinate Napoleon...
26. Re: Open source Euphoria...
DB James wrote:
>
> Jules wrote:
> >
> > DB James wrote:
> > >
> > > CChris wrote:
> > > >
> > > > Jules wrote:
> > > <SNIP>
> > >
> > > If changes in Eu code were done based on empirical evidence...
> >
> > Not sure what you mean by "empirical evidence"
>
> Derek or CChris or others could answer this better, but an example (kindly
> ignore errors) might be:
>
> Ima Coder loves the "enum" keyword from C and she wants to have a keyword in
> Eu called constantenum that works this way:
>
> constantenum Name=1,Adr1,Adr2,Adr3,Phone,Last --imagine this 50 items long
> sequence database,record
> record={}
> for i=1 to Last do record=append(record,{}) end for
> database=repeat(record,1000)
> database[1][Name]="Elvis Presley"
> database[1][Adr1]="123 Blue Suede Lane"
> ...
>
> Okay, so she implements this in a library such that when the code is
> processed, it turns the following:
>
> --constantenum Name=1,Adr1,Adr2,Adr3,Phone,Extra
>
> to:
>
> constant Name=1,Adr1=2,Adr2=3,Adr3=4,Phone=5,Extra=6
>
> People try it and report variously: "great!", "dumb", ,"unnecessary",etc.
>
> So, she gets serious, produces some code and puts it in a test version
> of Euphoria. She demonstrates that her code would save time in program
> development, etc. But some find it slows down the interpreter. She re-codes
> and now demonstrates that it is fast. But some might show that it would
> be better to simply borrow "enum" as used in C because (they may prove) it
> would be more flexible. And on like that until there is a general consensus
> to accept the code, change it, or forget it. (Or if the serious coders made
> the decision in actual practice, fine.)
>
> --Quark
ok, thanks. Maybe this will be of interest:
http://sern.ucalgary.ca/courses/cpsc/451/W98/Complexity.html
27. Re: Open source Euphoria...
Matt Lewis wrote:
>
> I did fix the file extension thing (nearly a one-liner--yay!). It was
> really a case of an incorrect error message.
OK, now I've fixed the memory leak. I'd updated EMalloc and friends in
be_alloc.c, but didn't fix FreeD in alloc.h. That's been updated, and
the SF bug page updated.
I also made a 3.1.1 branch in svn, and merged in all of the bugs that I
created^Wfixed (source only, I didn't check in any binary stuff, so if
you want a binary, you'll have to build it yourself).
Matt
28. Re: Open source Euphoria...
Jules wrote:
>
> DB James wrote:
> >
> > Jules wrote:
> > >
> > > DB James wrote:
> > > >
> > > > CChris wrote:
> > > > >
> > > > > Jules wrote:
> > > > <SNIP>
> > > >
> > > > If changes in Eu code were done based on empirical evidence...
> > >
> > > Not sure what you mean by "empirical evidence"
> >
> > Derek or CChris or others could answer this better, but an example (kindly
> > ignore errors) might be:
<SNIP>
> > the decision in actual practice, fine.)
> >
> > --Quark
>
> ok, thanks. Maybe this will be of interest: <a
> href="http://sern.ucalgary.ca/courses/cpsc/451/W98/Complexity.html">http://sern.ucalgary.ca/courses/cpsc/451/W98/Complexity.html</a>
No, not really. Perhaps you could summarize your view of the relevance of
that web doc to the issue at hand?
--Quark
29. Re: Open source Euphoria...
Matt Lewis wrote:
> Matt Lewis wrote:
>>
>> I did fix the file extension thing (nearly a one-liner--yay!). It was
>> really a case of an incorrect error message.
>
> OK, now I've fixed the memory leak. I'd updated EMalloc and friends in
> be_alloc.c, but didn't fix FreeD in alloc.h. That's been updated, and
> the SF bug page updated.
>
> I also made a 3.1.1 branch in svn, and merged in all of the bugs that I
> created^Wfixed (source only,
Thank you, Matt!
> I didn't check in any binary stuff, so if
> you want a binary, you'll have to build it yourself).
I'll try to do so, although I doubt that I'll succeed.
Regards,
Juergen
30. Re: Open source Euphoria...
> > ok, thanks. Maybe this will be of interest: <a
> > href="http://sern.ucalgary.ca/courses/cpsc/451/W98/Complexity.html">http://sern.ucalgary.ca/courses/cpsc/451/W98/Complexity.html</a>
>
> No, not really. Perhaps you could summarize your view of the relevance of
> that web doc to the issue at hand?
>
never mind, bad idea...
31. Re: Open source Euphoria...
Matt Lewis wrote:
>
> Matt Lewis wrote:
> >
> > I did fix the file extension thing (nearly a one-liner--yay!). It was
> > really a case of an incorrect error message.
>
> OK, now I've fixed the memory leak. I'd updated EMalloc and friends in
> be_alloc.c, but didn't fix FreeD in alloc.h. That's been updated, and
> the SF bug page updated.
>
> I also made a 3.1.1 branch in svn, and merged in all of the bugs that I
> created^Wfixed (source only, I didn't check in any binary stuff, so if
> you want a binary, you'll have to build it yourself).
OK. Great.
I guess in a few days I can start preparing a 3.1.1 release.
Regards,
Rob Craig
Rapid Deployment Software
http://www.RapidEuphoria.com
