headers
Daniel Jomphe | 1 Sep 2009 04:17
Picon

Re: Shen on github


Cool to see Shen on GitHub. This will definitely someday help people
collaborate on the project.

On gaining trust: way to go.

I've often pondered if the best Q&A site would be usable as a better
alternative to traditional forums. After all, any forum subject can be
posted in the form of a question. To this end, you may like
http://stackoverflow.com, which is better than any typical Q&A site,
and includes every feature we may want. All people have to do is to
make sure they tag their questions with "Qi" and/or "Shen". And the
biggest advantage is that questions and answers are both editable, so
that they can become better over time (or never become out of date).
Moreover, spam is not a problem there. One remark, though: it's
suggested not to use the site as a discussion board; it's more geared
towards answering questions than discussing and arguing. So you might
prefer to still have a forum / mailing list for this.

On Aug 31, 10:32 am, snorgers <stefan.ta...@...> wrote:
> I have created a repo on github called shen,
>
> http://github.com/snorgers/Shen/
>
> For anybody wanting to help developing, email me a message.
>
> I will use this standard, people who post code and fixes on the forum
> can gain a trust
> with me. When you feel comfortable please ask me about getting an
> account to help out hacking
(Continue reading)

Permalink | Reply | 5 comments |
headers
Daniel Jomphe | 1 Sep 2009 04:32
Picon

Re: core.qi


Wow, Stefan, your post and question very much highlight how much I
need to complete a second reading of the book before hoping to
understand your need and intention, and bring you useful help. I'm
sorry you spent the time to write this while I'm clearly not ready for
it. That said, I still commit to help on the development of Qi+Shen. I
will just start, let's say, doing much more easier stuff first. On a
positive note, my appetite to become useful at that level keeps
growing. So I might someday, not too far away, reach it...

On Aug 26, 5:45 pm, snorgers <stefan.ta...@...> wrote:
> Hi,
>
> This is a description of what I reasoned out of core.qi.
>
> 1. There should be a unifying notion of a define and a define lambda
> that unifies
> streaming unification matching parsing segmenting. The core function
> is reduce-help
> and it's here I'm working on getting all these components together,
> factoring out code,
> making it extensible. Although the code is growing, by  unifying
> concepts I expect the
> Qi-yacc code to shrink. also the mu reducer is going to communicate
> with this function
> in order to bring unificational pattern to a define and a define
> lambda
>
> 2. The define will be able to work on arbritrary streams and a
> protocol and tools for writing
(Continue reading)

Permalink | Reply | 4 comments |
headers
snorgers | 1 Sep 2009 07:22
Picon
Favicon

Re: core.qi


Also,

I skimmed through the jvm specification. Apparently there is gotos
in this specification. The question is how to take advantage of that.
using recur is a bit iniefficient. I'm thinking of patching clojure to
make
it have a tagbody like construct. Does anybody know if this is
possible?

/Stefan

On 1 Sep, 04:32, Daniel Jomphe <danieljom...@...> wrote:
> Wow, Stefan, your post and question very much highlight how much I
> need to complete a second reading of the book before hoping to
> understand your need and intention, and bring you useful help. I'm
> sorry you spent the time to write this while I'm clearly not ready for
> it. That said, I still commit to help on the development of Qi+Shen. I
> will just start, let's say, doing much more easier stuff first. On a
> positive note, my appetite to become useful at that level keeps
> growing. So I might someday, not too far away, reach it...
>
> On Aug 26, 5:45 pm, snorgers <stefan.ta...@...> wrote:
>
>
>
> > Hi,
>
> > This is a description of what I reasoned out of core.qi.
>
(Continue reading)

Permalink | Reply | 3 comments |
headers
Mark Tarver | 1 Sep 2009 13:36
Picon
Picon

Qilib as a discussion forum


This is something I had thought about and was going to mention - that
the task of maintaining this group will pass to my friend and
colleague, Dr Willi Riha.  He will follow my practice of only deleting
spam.  Hence this space will remain open and spam-free.

Mark

On 1 Sep, 03:17, Daniel Jomphe <danieljom...@...> wrote:
> Cool to see Shen on GitHub. This will definitely someday help people
> collaborate on the project.
>
> On gaining trust: way to go.
>
> I've often pondered if the best Q&A site would be usable as a better
> alternative to traditional forums. After all, any forum subject can be
> posted in the form of a question. To this end, you may likehttp://stackoverflow.com, which is better than
any typical Q&A site,
> and includes every feature we may want. All people have to do is to
> make sure they tag their questions with "Qi" and/or "Shen". And the
> biggest advantage is that questions and answers are both editable, so
> that they can become better over time (or never become out of date).
> Moreover, spam is not a problem there. One remark, though: it's
> suggested not to use the site as a discussion board; it's more geared
> towards answering questions than discussing and arguing. So you might
> prefer to still have a forum / mailing list for this.
>
> On Aug 31, 10:32 am, snorgers <stefan.ta...@...> wrote:
>
>
(Continue reading)

Permalink | Reply | 4 comments |
headers
Mark Tarver | 1 Sep 2009 13:37
Picon
Picon

Re: Qilang as a discussion forum


Sorry; that should have been Qilang of course.

On 1 Sep, 12:36, Mark Tarver <dr.mtar...@...> wrote:
> This is something I had thought about and was going to mention - that
> the task of maintaining this group will pass to my friend and
> colleague, Dr Willi Riha.  He will follow my practice of only deleting
> spam.  Hence this space will remain open and spam-free.
>
> Mark
>
> On 1 Sep, 03:17, Daniel Jomphe <danieljom...@...> wrote:
>
>
>
> > Cool to see Shen on GitHub. This will definitely someday help people
> > collaborate on the project.
>
> > On gaining trust: way to go.
>
> > I've often pondered if the best Q&A site would be usable as a better
> > alternative to traditional forums. After all, any forum subject can be
> > posted in the form of a question. To this end, you may likehttp://stackoverflow.com, which is better
than any typical Q&A site,
> > and includes every feature we may want. All people have to do is to
> > make sure they tag their questions with "Qi" and/or "Shen". And the
> > biggest advantage is that questions and answers are both editable, so
> > that they can become better over time (or never become out of date).
> > Moreover, spam is not a problem there. One remark, though: it's
> > suggested not to use the site as a discussion board; it's more geared
(Continue reading)

Permalink | Reply | 3 comments |
headers
Mark Tarver | 1 Sep 2009 14:03
Picon
Picon

Re: open space: discussion on Qi, Shen, licensing and open source


OK; this should have been clear from reading the license on the top of
the code.

 Beginning of Licence
;
; This software is licensed only for personal and educational use and
; not for the production of commercial software.  Modifications to
this
; program are allowed but the resulting source must be annotated to
; indicate the nature of and the author of these changes.
;
; Any modified source is bound by this licence and must remain
available
; as open source under the same conditions it was supplied and with
this
; licence at the top.

If you modify my code; you cannot put it under BSD.  And this applies
even with a commercial license which allows you to run Qi for money.
If you have a license then what you do with code written *on top* of
mine is your affair.  If you build something which runs under vanilla
Qi, and you have a license, you can close your code off or give it
away as you want.

Read the license and don't think that when I say you can do what you
want with 'your code' I am giving you carte blanche.  I am not.

***Hence if you really want to have Shen under BSD you would be
advised to port Shen to Clojure where you can experiment with your
(Continue reading)

Permalink | Reply | 0 comments |
headers
Henry Weller | 1 Sep 2009 14:31

Re: open space: discussion on Qi, Shen, licensing and open source


> ; This software is licensed only for personal and educational use and
> ; not for the production of commercial software.

What about for commercial use, e.g. writing a data manipulation tool to convert
files as part of a commercial project?  GPL would not preclude such a use
but it is not clear if your licence does.

> If you modify my code; you cannot put it under BSD.  And this applies
> even with a commercial license which allows you to run Qi for money.
> If you have a license then what you do with code written *on top* of
> mine is your affair.

Sure, I have bought your book and so I have a licence but it is not clear under
what conditions I can distribute my work as all of it appears to inherit your
licence.

> If you build something which runs under vanilla
> Qi, and you have a license, you can close your code off or give it
> away as you want.

That does not help Stefan; he needs the source code I have written and his code
is under the BSD licence.

> Read the license and don't think that when I say you can do what you
> want with 'your code' I am giving you carte blanche.  I am not.

It is still not clear how I can contribute to the Shen project if I cannot
distribute code needed for it under the licence it is to be distributed in.
(Continue reading)

Permalink | Reply | 0 comments |
headers
Mark Tarver | 1 Sep 2009 15:05
Picon
Picon

Re: open space: discussion on Qi, Shen, licensing and open source


The book license allows you to produce commercial work in Qi.  But it
does not extend to the Qi sources which remain under the original
license and this is explicitly stated in the license.

It's been there since November 2008.  I'm not springing any surprises.

I don't support FOSS or GPL so I feel no reason to conform to them.
I've written very publicly on why I don't accept them.

You can contribute to changing Qi sources if you want your changes to
remain under the Qi license.   Alternatively you can implement Shen in
Clojure and that's your work and you can put your license on it.
That's what Stefan was doing originally until a few days ago when the
course of development suddenly veered.

Mark

On 1 Sep, 13:31, Henry Weller <HWell...@...> wrote:
> > ; This software is licensed only for personal and educational use and
> > ; not for the production of commercial software.
>
> What about for commercial use, e.g. writing a data manipulation tool to convert
> files as part of a commercial project?  GPL would not preclude such a use
> but it is not clear if your licence does.
>
> > If you modify my code; you cannot put it under BSD.  And this applies
> > even with a commercial license which allows you to run Qi for money.
> > If you have a license then what you do with code written *on top* of
> > mine is your affair.
(Continue reading)

Permalink | Reply | 0 comments |
headers
Henry Weller | 1 Sep 2009 15:22

Re: open space: discussion on Qi, Shen, licensing and open source


> The book license allows you to produce commercial work in Qi.

Yes I understand but that is not the same as the example I gave: writing a data
manipulation tool to convert files for part of a commercial project.  Can such a
tool be used for a commercial project without buying you book or not?  This tool
is not distributed or included in a commercial product, only used for a
commercial project.

> You can contribute to changing Qi sources if you want your changes to
> remain under the Qi license.   Alternatively you can implement Shen in
> Clojure and that's your work and you can put your license on it.

But as you said you expect 80% of Shen is to be QiII in Qi source rather than
Clojure source, but the QiII sources are under your licence so how can they be
included in Shen if Shen is to be distributed under the BSD licence?

Henry
Permalink | Reply | 0 comments |
headers
Daniel Jomphe | 1 Sep 2009 15:50
Picon

Re: core.qi


Some time ago, I read an article or a blog post about how to emulate
gotos in languages which don't have it. In Java's case, the most
obvious use was using switch statements in a loop; and if I remember
well, each case (label) dispatched a function call instead of putting
all the "domain logic" code in the switch statement. Or it might have
ended up with no switch statement at all, just function calls. All of
this translates easily to Clojure, of course. That said, I don't know
about the performance costs of this approach, other than the fact that
the JVM's JIT compiler is very good with repetitive cases.

On Sep 1, 1:22 am, snorgers <stefan.ta...@...> wrote:
> Also,
>
> I skimmed through the jvm specification. Apparently there is gotos
> in this specification. The question is how to take advantage of that.
> using recur is a bit iniefficient. I'm thinking of patching clojure to
> make
> it have a tagbody like construct. Does anybody know if this is
> possible?
>
> /Stefan
>
> On 1 Sep, 04:32, Daniel Jomphe <danieljom...@...> wrote:
>
> > Wow, Stefan, your post and question very much highlight how much I
> > need to complete a second reading of the book before hoping to
> > understand your need and intention, and bring you useful help. I'm
> > sorry you spent the time to write this while I'm clearly not ready for
> > it. That said, I still commit to help on the development of Qi+Shen. I
(Continue reading)

Permalink | Reply | 2 comments |

Gmane