Name for a website about programming language design
James Harris
programming language design for any of us here to use. This post is to
ask for guidance on choosing a name.
First of all, what characteristics make a good name? The ones I can
think of are
* short
* easy to remember
* relating to the purpose
Any other desirable characteristics? Any existing established
precedents to follow?
Second, any specific suggestions for a name? (Whatever name is chosen
it will become *part* of a domain name. The rest of the domain name
will depend on the service used for hosting the site.)
James
BGB
'langdev' or 'langdesign' ?...
contents may also be an issue, like if/when set up, what sort of topics
to address and how to organize them.
in my own thinking, the merits of simplicity vs having lots of features
has been a thought, as I am presented with the issue that a lot of my
stuff is absurdly complex and unreasonably bulky, which is itself a
potential detractor...
after arguing some around here, it is notable that different people have
different perspectives and values on a lot of this.
admittedly, it also seems to tie in with some of my thoughts and
observations involving psychology, like how personality ties in with
coding, design, and aesthetic preferences, ... but this is a different
topic.
Rod Pemberton
news:1d8648f2-7b6b-4494...@k22g2000yqh.googlegroups.com...
code crud
syntax hax
semantics semantics
curly braces
LISP - Lost In Syntax Problems
IAC - It Ain't COBOL
parse this
HTH... ;-)
Rod Pemberton
Marco
any name that has some form of language and design in the name
Marco van de Voort
>> Any other desirable characteristics? Any existing established
>> precedents to follow?
>>
>> Second, any specific suggestions for a name? (Whatever name is chosen
>> it will become *part* of a domain name. The rest of the domain name
>> will depend on the service used for hosting the site.)
>>
>
> code crud
> syntax hax
> semantics semantics
> curly braces
> LISP - Lost In Syntax Problems
> IAC - It Ain't COBOL
> parse this
If we want to advocate strong typing:
- Bondage and Discipline
it also makes the artwork choice easier ( :-) )
BGB
but without some bondage how will one keep the newbies from getting
confused?...
the compiler needs to tie them to the bed and tell them they are a bad
boy for trying to implicitly convert between different integer types or
throwing or failing to catch an exception mentioned in a 'throws' clause...
now, as for syntax:
sadly, when a VM becomes non-trivial, the parser becomes the lesser of
ones' worries, as all the internal compiler and VM machinery becomes
much larger (large enough to make the relative complexity of the parser
irrelevant, at which point syntax becomes more of an aesthetic concern,
and the traditional "curly brace" syntax is generally regarded as most
aesthetically preferable AFAICT...).
and, even then, often in larger VMs, the VM itself becomes dwarfed
typically in library code and API wrappers.
so, maybe a few kloc for the parser, a few 10s of kloc for the compiler,
100s for VM and runtime support machinery, and several Mloc of class
libraries and glue code (at least this seems to be the case with J2SE...
OTOH, J2ME has a much smaller standard class library).
this latter point is why I am so insistent on having a good C interface:
most of this crap already exists in C and C++ land, and if one can start
leveraging some of it, then there is no need for massive class
libraries, and instead the class libraries can be focused on more core
tasks (and most of the rest can be via interfacing with C code or
similar...).
similarly, the class library can be more focused on providing interfaces
than on providing for specific use-cases, for example, providing a
generic class to perform file-IO tasks, rather than a big pile of
classes for every possible way a person might expect to make use of said
file.
but, alas, complexity is still an issue...
or such...
Mike Austin
All in all, I think a web site is great idea. BTW, how about we write a
language that looks like COBOL, but with curly braces and has the semantics of
LISP? <head esplodes>
But on a more serious note, what about a list of things we want to create a
site for prog langs? I think it would be really cool, we just have to figure
out what's different about it.
PiLuD has discussions and light wiki:
http://groups.google.com/group/pilud
C2 has dialog driven wiki:
http://c2.com/cgi/wiki
Lambda the Ultimate has good topics and discussions:
http://lambda-the-ultimate.org/
Programming Language News seems to be operating in the future??
http://plnews.org/
irc #proglangdesign is sometimes active.. it would be really cool if we all
hung out there :) There's also #ltu
So maybe it could be even higher level and these, as a starting point and
reference. BTW, I find the C2 wikis interesting, but they can sometimes turn
into flame wars or have incorrect information.
Mike
> HTH... ;-)
>
>
> Rod Pemberton
BGB
we need "MOAR JAVA!"...
(ok, not really...).
(bleh... it is sad, my newest language is largely based on one of the
major languages I have spent a good portion of my life despising...
grr... well, I only do it as I think I may be able to make something
less lame, at least for my own uses...).
> But on a more serious note, what about a list of things we want to
> create a site for prog langs? I think it would be really cool, we just
> have to figure out what's different about it.
>
> PiLuD has discussions and light wiki:
> http://groups.google.com/group/pilud
>
> C2 has dialog driven wiki:
> http://c2.com/cgi/wiki
>
> Lambda the Ultimate has good topics and discussions:
> http://lambda-the-ultimate.org/
>
> Programming Language News seems to be operating in the future??
> http://plnews.org/
>
damn that temporal continuity sometimes, always glitching out on us...
oh well, may as well use the opportunity to use one of those glass
touchscreen keyboards from Encom, and maybe a digitizing laser while I
am at it, then travel to the grid to use the time dilation for an
extreme coding speed boost...
and then one can wonder then why the login for Encom-OS says Ubuntu?...
"
BGB@Encom ~$ startlaser -now -ndstuff
IMA CHARGIN' MAH LAZER...
"
...
oh crap, "incomming game", stupid cubes from the sky...
> irc #proglangdesign is sometimes active.. it would be really cool if we
> all hung out there :) There's also #ltu
>
it is hard to talk about much of anything useful over either IRC or IM
in my experience, so I personally prefer to avoid it...
> So maybe it could be even higher level and these, as a starting point
> and reference. BTW, I find the C2 wikis interesting, but they can
> sometimes turn into flame wars or have incorrect information.
>
ok.
Tony
> From discussions last month ISTM worth setting up a website or wiki on
> programming language design for any of us here to use. This post is to
> ask for guidance on choosing a name.
What is the charter for the website or wiki? Is it an idea in the
formative/brainstorming stage? Is "it" (whatever "it" is) going to be
LIMITED TO design? What about specification, engineering and
implementation, etc.? Are those separate areas not to be covered by the
site? If you don't yet know WHAT it is, maybe naming it already is
premature, or if you simply MUST have a name now, call it FETUS. ;)
Rod Pemberton
news:oleVo.39164$Qi5....@newsfe01.iad...
> James Harris wrote:
> > From discussions last month ISTM worth setting up a website or wiki on
> > programming language design for any of us here to use. This post is to
> > ask for guidance on choosing a name.
>
> What is the charter for the website or wiki? Is it an idea in the
> formative/brainstorming stage? Is "it" (whatever "it" is) going to be
> LIMITED TO design? What about specification, engineering and
> implementation, etc.? Are those separate areas not to be covered by the
> site? If you don't yet know WHAT it is, maybe naming it already is
> premature, or if you simply MUST have a name now, call it FETUS. ;)
>
The name just needs to be something catchy or funny that's kind of related
and will be remembered.
Nobody who uses Twitter thinks of themselves as a twit, do they? But,
"twitter" is catchy...
Let's take the names of two operating system (OS) development websites:
OSDev.org
Bona Fide OS Developer
Which one is active? Which one is not?
They even get you to remember it's a .org site... It rolls right off the
tongue.
Let's consider the name of a site that claims to be:
"The world's leading source for pure technical x86 processor information."
It's called: Sandpile.org. It's easy to remember. It's a word associated
with an image. The name is not remotely related to the purpose of the
website. They don't even show a picture of a sandpile. It's just a name
you don't forget.
Rod Pemberton
Tony
I know, I get it: tailor it to the masses (especially the younger set who
have to text message each other every time nature calls) who apparently
are attracted to that kind of thing. I'd suck at mass-marketing because I
prefer things of substance rather than fluff. The whole slogan, cutsie
tootsie mascot thing, etc. does not appeal to me at all (Bright colors
and cartoon characters anyone? How about a free set of crayons if you
sign up?). High quality content is king, IMO. The rest simple does not
matter one ioda, zero, zilch (to me, and it is even a negative).
I didn't catch the result of last week's posts about this topic. Has a
charter been drafted? Who are the target participants? Has the "value
add" over existing avenues been established? Is feasible? What is the
likelihood of success and what are the criteria to judge "success" by? Is
"it" just a non-serious, ad-hoc, bottom-up thing to do just for the fun
of it?
Lots of questions... a name? Ech. Task #52 on the project plan: establish
name for "it".
Marco van de Voort
> this latter point is why I am so insistent on having a good C interface:
> most of this crap already exists in C and C++ land, and if one can start
> leveraging some of it, then there is no need for massive class
> libraries, and instead the class libraries can be focused on more core
> tasks (and most of the rest can be via interfacing with C code or
> similar...).
There are multiple levels of this:
- being C in the expected versions
- being a different C (still being able to "mostly" read headers)
- be a non C procedural language that can be mapped to C (our case)
But anyway, I don't really believe in this kind of leveraging. Usually so
much of the originating system bleeds through, you might as well use the
originating system directly.
James Harris
> On 1/5/2011 2:08 AM, James Harris wrote:
>
>
>
> > From discussions last month ISTM worth setting up a website or wiki on
> > programming language design for any of us here to use. This post is to
> > ask for guidance on choosing a name.
>
> > First of all, what characteristics make a good name? The ones I can
> > think of are
>
> > * short
> > * easy to remember
> > * relating to the purpose
>
> > Any other desirable characteristics? Any existing established
> > precedents to follow?
>
> > Second, any specific suggestions for a name? (Whatever name is chosen
> > it will become *part* of a domain name. The rest of the domain name
> > will depend on the service used for hosting the site.)
>
> 'langdev' or 'langdesign' ?...
I think your suggestions may be better but just to throw a couple of
other suggestions into the mix
pld - for programming language design
pldi - for programming language design and implementation
James
James Harris
...
> All in all, I think a web site is great idea. BTW, how about we write a
> language that looks like COBOL, but with curly braces and has the semantics of
> LISP? <head esplodes>
Haha - sounds like the beginning of an essay on what not to do.
procedure section {
add (idea) to (similar-ideas).
move (similar-ideas) to bin.
}
> But on a more serious note, what about a list of things we want to create a
> site for prog langs? I think it would be really cool, we just have to figure
> out what's different about it.
ISTM that some of the discussions on comp.lang.misc result in really
insightful conclusions but the findings can get a bit swamped in the
rest of the discussion. A web site would be a great place to write
them.
...
> irc #proglangdesign is sometimes active.. it would be really cool if we all
> hung out there :) There's also #ltu
AIUI IRC it is for immediate discussion: stuff typed appears
immediately - or at least once Enter is pressed - and different
conversations get interleaved. Does the transcript get saved
somewhere?
A big plus about Usenet is that conversations can be referred to later
and conversations are not interleaved - at least not quite so much.
> So maybe it could be even higher level and these, as a starting point and
> reference. BTW, I find the C2 wikis interesting, but they can sometimes turn
> into flame wars or have incorrect information.
Robbert had some ideas about organising content. For flame wars we
always have Usenet.
Maxims and joint conclusions could go on the web site unattributed but
because there will be different ideas how about encouraging anyone who
writes a personal article for the web site to add his name to it? Then
he or she gets the credit and anyone reading can see that it's a
personal opinion.
James
James Harris
> Will it be a Wiki? If so it should be easy to use and join but also easy for the moderator to "undo" poor edits or spam (unfortunately the whole WWW is in need of adult supervision - sigh )
Um, yes, I know that can be done with a wiki. Not sure about a web
site. (I can see advantages to both but only have experience with a
wiki service.) Maybe depends on the facilities offered by the specific
hosting service chosen. Something I'll come back to.
> any name that has some form of language and design in the name
OK
James
James Harris
> James Harris wrote:
> > From discussions last month ISTM worth setting up a website or wiki on
> > programming language design for any of us here to use. This post is to
> > ask for guidance on choosing a name.
>
> What is the charter for the website or wiki? Is it an idea in the
> formative/brainstorming stage? Is "it" (whatever "it" is) going to be
> LIMITED TO design? What about specification, engineering and
> implementation, etc.? Are those separate areas not to be covered by the
> site?
I think we're still open to suggestions. There was a discussion which
you can see at
https://groups.google.com/group/comp.lang.misc/browse_frm/thread/93501b369a148fd6?hl=en#
As far as I'm concerned almost anything related to creating a
programming language or to what we try to do on comp.lang.misc is
relevant - including all the activities you mention. Though that's
just an idea; different views are welcome.
James
James Harris
wrote:
...
> Let's take the names of two operating system (OS) development websites:
>
> OSDev.org
> Bona Fide OS Developer
>
> Which one is active? Which one is not?
>
> They even get you to remember it's a .org site... It rolls right off the
> tongue.
I think both are pretty good sites. I suppose osdev is a good name in
that it is short and easy to remember while also completely
encapsulating the principle of what it is about.
>
> Let's consider the name of a site that claims to be:
> "The world's leading source for pure technical x86 processor information."
>
> It's called: Sandpile.org. It's easy to remember. It's a word associated
> with an image. The name is not remotely related to the purpose of the
> website. They don't even show a picture of a sandpile. It's just a name
> you don't forget.
Interesting points, Rod. Do you think that maybe the above are easy to
remember partly because there are few or no other similar sites?
There seem to be very many web sites on programming language design
(but few or none where our edits would be given much visibility; most
don't accept any updates).
As a test for names, try one search for aodfaq and another for
codewiki. They come up near the top of the list when I try either a
Yahoo search or a Google search while not logged in to a Google
account. (I think Google customises results for logged-in accounts so
it may give a higher rank to pages related to what one has searched
for before. Not sure if they customise them based on cookies too.)
Both those names were set up as sites on wikispaces fairly recently -
in the last few years - so it's gratifying that they appear on the
first page of search results.
James
James Harris
> Rod Pemberton wrote:
...
> > Let's consider the name of a site that claims to be:
> > "The world's leading source for pure technical x86 processor
> > information."
>
> > It's called: Sandpile.org. It's easy to remember. It's a word
> > associated with an image. The name is not remotely related to the
> > purpose of the website. They don't even show a picture of a
> > sandpile. It's just a name you don't forget.
>
> I know, I get it: tailor it to the masses (especially the younger set who
> have to text message each other every time nature calls) who apparently
> are attracted to that kind of thing. I'd suck at mass-marketing because I
> prefer things of substance rather than fluff. The whole slogan, cutsie
> tootsie mascot thing, etc. does not appeal to me at all (Bright colors
> and cartoon characters anyone? How about a free set of crayons if you
> sign up?). High quality content is king, IMO. The rest simple does not
> matter one ioda, zero, zilch (to me, and it is even a negative).
Yes, content is king. Information is usually more useful than
crayons. :-)
> I didn't catch the result of last week's posts about this topic.
You do know about Google's archive, no? The thread is still available
at
http://groups.google.com/group/comp.lang.misc/browse_frm/thread/93501b369a148fd6?hl=en#
> Has a
> charter been drafted? Who are the target participants? Has the "value
> add" over existing avenues been established? Is feasible? What is the
> likelihood of success and what are the criteria to judge "success" by? Is
> "it" just a non-serious, ad-hoc, bottom-up thing to do just for the fun
> of it?
>
> Lots of questions... a name? Ech. Task #52 on the project plan: establish
> name for "it".
A charter: No. Comments on topicality are welcomed still at
http://groups.google.com/group/comp.lang.misc/browse_frm/thread/93501b369a148fd6?hl=en#
Target participants: users of comp.lang.misc and anyone else who wants
to be involved on the topics related to language design. I don't
expect there to be a lot of writers but there may be many readers,
depending on what gets added.
Value add: see the earlier discussion.
Feasible: yes, see the earlier discussion, though a hosting service
has not been chosen yet.
Success: IMHO it will be successful if it's useful to *us*. I already
have some things to use it for. As with programming languages a good
principle is: design for yourself to use. If others like it then that
just adds to its success but it's not intended to make money or
achieve any other aim than to be useful.
For your last question: not it's not for fun. I have no overriding
preference for a specific approach but I think it's most likely to be
bottom up for the simple reason that for a subject as large as
programming language development a top-down approach can become a
massive burden. With bottom-up one begins with useful content and adds
to it as time and interest permits.
All I've written above are just suggestions. Disagreement and other
ideas are welcome.
James
Jacko
As for names? ummm....
parsintime.org
langpimp.org
askibot.org
Cheers Jacko
BGB
well, it depends on the reasons for creating a new language...
if a person doesn't have a problem with C, and has no intention of
trying to replace C, then having a good C interface makes sense.
more so if both languages are C-family anyways...
James Harris
wrote:
...
> OSDev.org
Rod's post must have been in the back of my mind. Have just thought of
another suggestion for the name, derived in a not-so-subtle way from
the above:
pldev
It is nice and short (and meaningful?) and has a surprising advantage
that there are apparently few such words out there. Google currently
reports only 16,500 hits.
Just a suggestion.
James
Tony
I was hoping for a summary that showed progress, results, plans,
conclusions or something, but it's OK.
OK. I think you've told me (indirectly) exactly what I wanted to know.
Have fun.
Even if it is just a club for an existing group of people or for
hobbyists though, be aware that the ownership of content (which looks
like it will come from participants) and the terms of agreement of the
usage of such content now and in perpetuity is very, very important to
have documented clearly, even if such terms are implicit from governing
laws, and that they be understood by all regular participants and
transients. There's a not-so-well-known-by-the-uninitiated strategy that
exploits the many in favor of a few via "games" played with cleverly
worded or implicit terms, and you probably want to avoid giving any
incling of such trickery, so it behooves you to be very diligent in
developing the terms of agreement and remembering that what isn't there
is as important as what is, if not moreso.
Robbert Haarman
On Sun, Jan 09, 2011 at 09:47:29AM -0800, James Harris wrote:
>
> pldev
>
> It is nice and short (and meaningful?) and has a surprising advantage
> that there are apparently few such words out there. Google currently
> reports only 16,500 hits.
I like it. As it happens, I also run a small hosting service, which puts
me in a position where I can cheaply register domain names. Shall I go
ahead and register pldev.net? pldev.org also seems to be available,
as does pldev.com.
Cheers,
Bob
James Harris
wrote:
...
> I like it. As it happens, I also run a small hosting service, which puts
> me in a position where I can cheaply register domain names. Shall I go
> ahead and register [...]
Thanks for the offer. Maybe hold off regarding the final name (unless
you want one yourself, of course, or just want to reserve them). ISTM
the next steps are
* finish up the discussion of a base name and, as Tony has pointed
out, start talking about licensing
* discuss hosting solutions - post about that next week. Would be a
good time to discuss and agree the final name including such as you
mentioned above
* get the site started properly maybe around the end of January or
beginning of February.
Just a suggested outline. I'd like to make sure everyone has a chance
to comment and have some personal time pressures up to the end of this
month so I hope the above is reasonable. There's room for anything
missing to be added.
James
James Harris
...
> Even if it is just a club for an existing group of people or for
> hobbyists though, be aware that the ownership of content (which looks
> like it will come from participants) and the terms of agreement of the
> usage of such content now and in perpetuity is very, very important to
> have documented clearly, even if such terms are implicit from governing
> laws, and that they be understood by all regular participants and
> transients.
Tony has raised a good point. It is important to know what licence
website content will be published under. Any suggestions?
Do we need to distinguish the terms for written articles from the
terms for source code? And maybe other types?
AIUI, regardless of licensing terms, ownership and *copyright* is
generally retained by the author (unless employed to write in which
case copyright may be held by the employer). I think of licensing as
relating to what people can do with the work.
James
BGB
maybe GFDL or Creative-Commons ShareAlike?...
did find this:
http://en.wikipedia.org/wiki/GFDL
http://en.wikipedia.org/wiki/Share-alike
Tony
> On Jan 12, 2:00 pm, Robbert Haarman <comp.lang.m...@inglorion.net>
> wrote:
>
> ...
>
>> I like it. As it happens, I also run a small hosting service, which
>> puts me in a position where I can cheaply register domain names.
>> Shall I go ahead and register [...]
>
> Thanks for the offer. Maybe hold off regarding the final name (unless
> you want one yourself, of course, or just want to reserve them). ISTM
> the next steps are
>
> * finish up the discussion of a base name and, as Tony has pointed
> out, start talking about licensing
I was trying to focus you on "Terms of Service". For example, what
happens to a registered (assuming one will be able to register, i.e.)
participant's postings and such when they unregister? What about
transient participant's posts and such? Will they be able to remove them?
Modify them? Etc. The following kind of thing I find unacceptable (read,
I won't register with a site that has any such TOS):
"By posting content or making content available for inclusion on the
<site name> forums, including without limitation threads and comments you
post to the forum, you grant <site name> and its proprietors a
world-wide, perpetual, irrevocable, royalty free, non-exclusive, fully
sub-licensable license to use, distribute, reproduce, modify, adapt,
publish, translate, publicly perform and publicly display, and to
incorporate such content into other works in any format or medium now
known or later developed."
Rod Pemberton
Is this a can of worms? Everyone has their own preference... Countries
have different laws...
Obviously, there is a large GPL/LGPL license crowd. Besides them, there is
a smaller BSD license crowd. Are there any other licenses that are widely
used today? I.e., not just for one important project? I'm not aware of any
others with "critical mass". MIT license was used years ago.
If the license is incompatible with GPL, will it be used? If it's for
something "trivial" and standalone, say a bootsector, I think it would.
But, if it's something larger, people are going to want to link it to other
stuff, and modify it, and apply their choice of license to their changes (
... which may or may not be legal. Supposedly, copyright applies to the
work as a whole, not to trivial changes. So, that's another issue.) So, if
you deny the GPL, is the code effectively "dead"? If you choose another
well known license, say BSD or MIT, will the code be used? What if you
choose a license that few people have heard of? E.g., let's say you choose
the Apache License, will that discourage some from using the code? I think
it will.
I like Public Domain (no copyrights or copyrights terminated), but that may
or may not be legal everywhere. Some sources say it is due to treaties, but
other sources say it's not because some countries have no legal version of
non-copyrighted works, i.e., everything is copyrighted including works that
pre-existed copyright law. 2-clause BSD is about the closest I think one
can come to PD with licensed code. Of course, if it's copyrighted and
licensed, it can be licensed under different licenses, AIUI. That may be
important. I.e., it's possible the site could license the code as GPL and
BSD, and by user request, etc. Then, the code has more flexibility as to
who uses it. AFAIK, the copyright owner retains ownership and copyrights
(at least US law AIUI...), even if one is allowed to license the work with
all rights. I think that's the reason why I've seen some open-source
projects become closed source and commercial.
I've been meaning to obtain the legal answers (for the US) to some of those
issues to help me decide how to release some of my code, but haven't yet and
probably won't for some time. You might look at OSI. They've reviewed most
open source licenses. If I didn't miscount, there are 66 types of licenses.
They have a slightly different legal perspective than the FSF as to what
open source is. They have 10 items they use to classify a license as open
source. Unfortunately, I think the GPL restricts a couple of those items...
IMO, GPL is not compatible with their list! So, I'm not sure how that is
resolved when mixing GPL code with different licenses. I.e., does mixing
GPL with another license place GPL restrictions on the other code? Or,
vice-versa? Since I've not looked through all 66 of them, there might be
other simple licenses like 2-clause BSD, but they probably aren't well
known.
Rod Pemberton
James Harris
wrote:
...
> Do we need to distinguish the terms for written articles from the
> terms for source code? And maybe other types?
>
> AIUI, regardless of licensing terms, ownership and *copyright* is
> generally retained by the author (unless employed to write in which
> case copyright may be held by the employer). I think of licensing as
> relating to what people can do with the work.
Having thought a bit more about the licensing (and read the responses)
ISTM that a site with an arbitrary number of contributors doesn't have
to mandate one licence for everyone's work. It could have a default
licence that suits most people and anyone who wanted to write
something under a different licence could include text to say so. The
default licence could be simple and thus avoid a lot of what are
evidently quite complex issues.
A couple of other thoughts:
I don't mind my name appearing on my own work but if someone copies
and changes my work I think I'd rather my name was taken off - so
unless I've had a chance to review the changes I think I'd like to
avoid a licence that requires attribution.
While anything published is intended to be helpful no author wants to
be held liable for it or how it is used.
So how about making the *default* licence the following for all
content: text, source code and anything else?
http://creativecommons.org/publicdomain/zero/1.0/
Compared to the other licences - especially the horrendous Gnu
licences - it is blissfully simple yet addresses the issues above.
Further, I wondered what licence was used on osdev which has code and
articles and found they conducted a poll for each:
http://wiki.osdev.org/OSDev_Wiki:About
http://forum.osdev.org/viewtopic.php?f=8&t=13841&hilit=poll
http://forum.osdev.org/viewtopic.php?f=8&t=13840&hilit=poll
You can see that, although few people voted, in both cases (the latter
two links) public domain was more popular.
Maybe there are better options...?
James
Tony
"Rod Pemberton" <do_no...@notreplytome.cmm> wrote in message
news:ign309$16r$1...@speranza.aioe.org...
FWIW, I won't touch (or look at) viral GPL code. (No comments here please
unless you must, because I know this is flamewar topic).
I'm not sure what JH was on about already with the license thing anyways.
You seem to know though. What am I not "getting"? What is he looking to
license?
Tony
> On Jan 12, 10:14 pm, James Harris <james.harri...@googlemail.com>
> wrote:
>
> ...
>
>> Do we need to distinguish the terms for written articles from the
>> terms for source code? And maybe other types?
>>
>> AIUI, regardless of licensing terms, ownership and *copyright* is
>> generally retained by the author (unless employed to write in which
>> case copyright may be held by the employer). I think of licensing as
>> relating to what people can do with the work.
>
> Having thought a bit more about the licensing (and read the responses)
> ISTM that a site with an arbitrary number of contributors doesn't have
> to mandate one licence for everyone's work.
Ah, OK, NOW I get what were "on" about. I was thinking you meant that,
but wasn't sure. Yes, look at Codeproject for example. It does the
"upload it with any license you want and let the reader figure it out".
It's easy to do one better though: allow searching and such by license in
addition to other ways. For instance, I don't care to see anything
associated with GPL.
> It could have a default
> licence
I don't see why that is good. It sounds rather bizarre to me actually.
> that suits most people and anyone who wanted to write
> something under a different licence could include text to say so. The
> default licence could be simple and thus avoid a lot of what are
> evidently quite complex issues.
Something simpler: don't worry about it at all. Just worry about how you
want your own stuff licensed and accomodate all others.
Clearly you have some greater notions of what this thing you are building
is. I'm trying to figure out how you got from NNTP newsgroup (this one)
to whatever the ideas in your mind are now that requires licensing to be
discussed. Obviously this discussion made me think of Codeproject, but
that seems like something so completely different than this newsgroup and
any related upstart derived from it on the web. Do tell.
>
> A couple of other thoughts:
>
> I don't mind my name appearing on my own work but if someone copies
> and changes my work I think I'd rather my name was taken off - so
> unless I've had a chance to review the changes I think I'd like to
> avoid a licence that requires attribution.
Again, make a license for your stuff as you desire. Again though, what
are you considering licensing? I think of source code when you say
"license". But you said "content" before. Please explain your thoughts
and comments.
>
> While anything published is intended to be helpful no author wants to
> be held liable for it or how it is used.
>
> So how about making the *default* licence the following for all
> content: text, source code and anything else?
Bad, bad, bad. Please get the idea of "default license" out of your head.
Also, I think you are still looking for a TOS and not a license. Note
that IANAL, and anything I say or do, will be held against you (read, if
I am bad, it is YOUR fault). ;)
>
> http://creativecommons.org/publicdomain/zero/1.0/
>
> Compared to the other licences - especially the horrendous Gnu
> licences - it is blissfully simple yet addresses the issues above.
>
> Further, I wondered what licence was used on osdev which has code and
> articles and found they conducted a poll for each:
>
> http://wiki.osdev.org/OSDev_Wiki:About
> http://forum.osdev.org/viewtopic.php?f=8&t=13841&hilit=poll
> http://forum.osdev.org/viewtopic.php?f=8&t=13840&hilit=poll
>
> You can see that, although few people voted, in both cases (the latter
> two links) public domain was more popular.
>
> Maybe there are better options...?
Surely. But what are you thinking in the way of source code other than
non-actual-product snippets-for-example in discussion? I thought "it" was
going to be a discussion group or something?
(Geez Louise, I don't read a few posts and I'm lost? Are you "hell-bent"
on NOT doing this project as a project (with planning, process, etc.)?
Rod Pemberton
>
> So how about making the *default* licence the following for all
> content: text, source code and anything else?
>
> http://creativecommons.org/publicdomain/zero/1.0/
>
I like that it says PD. But...
You're aware that link you posted is to the "license deed" for what
CreativeCommons is calling CC0 or "no rights reserved", yes? It's not the
link for the actual CC0 license. The CC0 license is here:
http://creativecommons.org/publicdomain/zero/1.0/legalcode
Did you read that? It's supposed to be the license which is equivalent to
the deed... Make sure you have ten minutes. :-)
Is the CC0 license in the OSI list? I don't see it by name. Actually, I
don't see the other CreativeCommons licenses either, although there are two
with "Common" in their titles. If not reviewed by OSI, license
compatibility could be an issue, if that matters.
OSI does list 9 "major" licenses currently in use:
http://opensource.org/proliferation-report
They declared the rest redundant, etc. Of the 9, BSD and MIT are the
simplest. Other simple ones are Fair License, zlib/libpng, NTP, ISC,
Eiffel, Boost Software. Most of those are probably equivalent to BSD/MIT.
One or two looked simpler. None are PD though.
Rod Pemberton
BGB
the report is mostly dealing with source-code licenses, whereas the
presumed topic was for documentation/text licenses, where things like
GFDL and Creative Commons licenses are more common...
many source-code licenses are essentially ambiguous when applied to
documentation, given documentation is created and used differently than
source code.
in a typical program, the documentation is "along for the ride", but
matters differ when a work is primarily or solely documentation.
Rod Pemberton
news:igpt5i$pcc$1...@news.albasani.net...
Oh...
I presumed that programming related website would have a large amount of
code content, e.g., 30 to 70%, as compared to other websites. It's possible
that someone would want to take the non-code content or documentation and
use it too. In that case, documentation license would be a better choice.
But, I'm still of the perspective that the code will or could be copied and
used by others far more frequently. So, IMO, the code content should be
under a useable code license.
Rod Pemberton
BGB
well, that is possible...
also possible is that it ends up being a large collection of mostly
theoretical information (where people do the whole academic thing and
write about the Hindley-Milner type-system using lots of set-notation
and typed lambda-calculus, which then consequently almost no one besides
the original author can make sense of it).
hopefully though, practical information could be present, like possible
ways of designing and implementing things and their relative trade-offs, ...
but, if it ends up being mostly code, then it is not so much about VM
and language design, so much as it would be an informal VM / language
implementation.
or such...
Jacko
latest. It's an quite a lot done Java ME version of Tcl about version
8.4, without the word 'socket' and file urls. Makes me wonder if
socket has a high level definition in terms of files and threaded
blocking.
I've learnt more about tcl doing this, and will add a much simpler
than tcl blend java class user.
Hopefully optimization will keep the file size small.
But both info and code would be good. Open licence of some sort.
Cheers Jacko
Marco van de Voort
> I like Public Domain (no copyrights or copyrights terminated), but that may
> or may not be legal everywhere. Some sources say it is due to treaties, but
> other sources say it's not because some countries have no legal version of
> non-copyrighted works, i.e., everything is copyrighted including works that
> pre-existed copyright law.
Well, I do know that many continental European states do know PD, but only
as term for works whose copyright has expired. You can't "put" a work into
PD as some common law countries afaik allow, copyright is inalienable, and
rights can't be signed away.
> I've been meaning to obtain the legal answers (for the US) to some of those
> issues to help me decide how to release some of my code, but haven't yet and
> probably won't for some time. You might look at OSI. They've reviewed most
> open source licenses. If I didn't miscount, there are 66 types of licenses.
> They have a slightly different legal perspective than the FSF as to what
> open source is. They have 10 items they use to classify a license as open
> source. Unfortunately, I think the GPL restricts a couple of those items...
> IMO, GPL is not compatible with their list! So, I'm not sure how that is
> resolved when mixing GPL code with different licenses. I.e., does mixing
> GPL with another license place GPL restrictions on the other code? Or,
> vice-versa? Since I've not looked through all 66 of them, there might be
> other simple licenses like 2-clause BSD, but they probably aren't well
> known.
Keep in mind that many people that think that GPL is grosso modo a good
thing, often think it should only be used in strategic cases.
People should also not forget that the three posterchildren for GPL are "special":
1. FSF (GNU) projects require signing away rights to the FSF foundation,
which can relicense if necessary (e.g. for transitions like GPL2->3). Yes
you need to mail a signed declaration for every non trivial change.
2. Linux kernel is a kernel, as in that no apps link to it in the GPL sense,
and the GPL carries provisions for the "OS".
3. Debian; Debian is a distribution and mostly produces "metadata", not real code.
IMHO keep BSD. Or if it is a library and the license must smell like a GNU,
GPL with static linking exception. (which, despite its name, is lighter in
requirements than even L-GPL)
James Harris
...
> Again, make a license for your stuff as you desire. Again though, what
> are you considering licensing? I think of source code when you say
> "license". But you said "content" before. Please explain your thoughts
> and comments.
Have you read the earlier thread I referred you do? It would help if
you could raise your points there - to keep the discussion about
content together.
> > While anything published is intended to be helpful no author wants to
> > be held liable for it or how it is used.
>
> > So how about making the *default* licence the following for all
> > content: text, source code and anything else?
>
> Bad, bad, bad. Please get the idea of "default license" out of your head.
> Also, I think you are still looking for a TOS and not a license. Note
> that IANAL,
Not a lawyer, OK.
> and anything I say or do, will be held against you (read, if
> I am bad, it is YOUR fault). ;)
Sounds like some women I know. :-) No offence to female-kind in
general. Only *some* women.
> > http://creativecommons.org/publicdomain/zero/1.0/
>
> > Compared to the other licences - especially the horrendous Gnu
> > licences - it is blissfully simple yet addresses the issues above.
>
> > Further, I wondered what licence was used on osdev which has code and
> > articles and found they conducted a poll for each:
>
> > http://wiki.osdev.org/OSDev_Wiki:About
> > http://forum.osdev.org/viewtopic.php?f=8&t=13841&hilit=poll
> > http://forum.osdev.org/viewtopic.php?f=8&t=13840&hilit=poll
>
> > You can see that, although few people voted, in both cases (the latter
> > two links) public domain was more popular.
>
> > Maybe there are better options...?
>
> Surely. But what are you thinking in the way of source code other than
> non-actual-product snippets-for-example in discussion? I thought "it" was
> going to be a discussion group or something?
No, not a discussion group. That's what Usenet is for.
James
James Harris
wrote:
> "James Harris" <james.harri...@googlemail.com> wrote in message
...
> > So how about making the *default* licence the following for all
> > content: text, source code and anything else?
>
> > http://creativecommons.org/publicdomain/zero/1.0/
>
> I like that it says PD. But...
>
> You're aware that link you posted is to the "license deed" for what
> CreativeCommons is calling CC0 or "no rights reserved", yes? It's not the
> link for the actual CC0 license. The CC0 license is here:http://creativecommons.org/publicdomain/zero/1.0/legalcode
>
> Did you read that? It's supposed to be the license which is equivalent to
> the deed... Make sure you have ten minutes. :-)
Yes. The legal version and the user-friendly version should carry the
same thoughts. The main difference, ISTM, is that the friendly version
includes a specific comment that people should not imply endorsement.
I guess that's out-with the remit of the CC0 licence itself but is a
good clause to have.
Note the answer to the question "What is the difference between CC0
and the Public Domain Mark?" on
http://wiki.creativecommons.org/CC0_FAQ
"CC0 and PDM differ in important respects and have distinct purposes.
CC0 is intended for use only by authors or holders of copyright and
related or neighboring rights (including sui generis database rights),
in connection with works that are still subject to those rights in one
or more jurisdictions. PDM, on the other hand, can be used by anyone,
and is intended for use with works that are already free of known
copyright restrictions throughout the world. The tools also differ in
terms of their effect when applied to a work. CC0 is legally operative
in the sense that when it is applied, it changes the copyright status
of the work, effectively relinquishing all copyright and related or
neighboring rights worldwide. PDM is not legally operative in any
respect – it is intended to function as a label, marking a work that
is already free of known copyright restrictions."
In other words, CC0 is specifically for the owner of a work to use.
>
> Is the CC0 license in the OSI list? I don't see it by name. Actually, I
> don't see the other CreativeCommons licenses either, although there are two
> with "Common" in their titles. If not reviewed by OSI, license
> compatibility could be an issue, if that matters.
>
> OSI does list 9 "major" licenses currently in use:http://opensource.org/proliferation-report
>
> They declared the rest redundant, etc. Of the 9, BSD and MIT are the
> simplest. Other simple ones are Fair License, zlib/libpng, NTP, ISC,
> Eiffel, Boost Software. Most of those are probably equivalent to BSD/MIT.
> One or two looked simpler. None are PD though.
I don't really know what the OSI list is for. As you say, they don't
seem to have a declaration of public domain.
James
James Harris
> From discussions last month ISTM worth setting up a website or wiki on
> programming language design for any of us here to use. This post is to
> ask for guidance on choosing a name.
>
> First of all, what characteristics make a good name? The ones I can
> think of are
>
> * short
> * easy to remember
> * relating to the purpose
>
> Any other desirable characteristics? Any existing established
> precedents to follow?
>
> Second, any specific suggestions for a name? (Whatever name is chosen
> it will become *part* of a domain name. The rest of the domain name
> will depend on the service used for hosting the site.)
The names put forward were
langdev
langdesign
code crud
syntax hax
semantics semantics
curly braces
LISP - Lost In Syntax Problems
IAC - It Ain't COBOL
parse this
Bondage and Discipline
fetus
pld
pldi
parsintime
langpimp
askibot
pldev
pldevel
Apart from being amusing I don't think names with spaces are really
practical. Of the rest the best seem to me to be pldev, langdev and
pldi in that order (best to worst). Robbert has already indicated a
preference for pldev. What does everyone else think?
One thing to think about: if this is to go on google sites (which is
by no means decided) we would need a name that is 6 chars or more.
That's why I've added pldevel at the bottom.
A separate post about hosting is to follow later.
James
Mike Austin
I like langdev, langdesign, and pldev (I guess that makes me pragmatic), but
saying pldevel is harder to say and sounds French :) piele-devele
Well, I guess instead of language by committee, we got web site by committee.
It's a good experience.
> James
Marco van de Voort
(Well, I meant B&D more as a sub-category, rather than the forum name)
I like "parse this" the most (even despite the fact that language design is
not just parsing).
> Apart from being amusing I don't think names with spaces are really
> practical.
Hyphenated domainnames are allowed afaik?
>Of the rest the best seem to me to be pldev, langdev and
> pldi in that order (best to worst). Robbert has already indicated a
> preference for pldev. What does everyone else think?
langdev then. At least it is directly obvious, I had to think deep about pl*
James Harris
wrote:
...
> langdev then. At least it is directly obvious, I had to think deep about pl*
OK. I guess the pl* names look familiar to me due to
http://www.sigplan.org/pldi.htm
Human languages can be developed too. Doesn't langdev seem to allow
for human language development?
James
James Harris
>
> On Sun, Jan 09, 2011 at 09:47:29AM -0800, James Harris wrote:
>
> > pldev
>
> > It is nice and short (and meaningful?) and has a surprising advantage
> > that there are apparently few such words out there. Google currently
> > reports only 16,500 hits.
>
> I like it. As it happens, I also run a small hosting service, which puts
> me in a position where I can cheaply register domain names. Shall I go
Hi Bob,
I sent you an e-mail a few days ago on this topic. Did you get it? I
may have the wrong e-mail address.
James
Marco van de Voort
> OK. I guess the pl* names look familiar to me due to
>
> http://www.sigplan.org/pldi.htm
>
> Human languages can be developed too. Doesn't langdev seem to allow
> for human language development?
I'm sorry? Did the question change? Do you want the best name, or just the
name with the least confusion? :-)
Domain matching is based on first order gut feeling. First order gut
feelings are never 100% correct, they are just a decent match with minimal
investment.
Anyway, in retrospect, I think "langdesign" is even better. I don't see the
use of the pl* acronyms at all, since you only "get" them after you learned
the meaning, which kind of defeats the purpose.
Robbert Haarman
>
> Hi Bob,
>
> I sent you an e-mail a few days ago on this topic. Did you get it? I
> may have the wrong e-mail address.
Hi James,
Thanks for reminding me. I did indeed receive the email, but I had
overlooked it until now. I'm feeling a bit sick at the moment, but I
will get back to you once I'm feeling better.
Cheers,
Bob
James Harris
wrote:
> On Fri, Jan 21, 2011 at 03:03:14AM -0800, James Harris wrote:
>
> > Hi Bob,
>
> > I sent you an e-mail a few days ago on this topic. Did you get it? I
> > may have the wrong e-mail address.
>
> Hi James,
>
> Thanks for reminding me. I did indeed receive the email, but I had
> overlooked it until now. I'm feeling a bit sick at the moment,
Sorry you're not feeling too well.
> but I
> will get back to you once I'm feeling better.
OK thanks. No hurry.
James
Rod Pemberton
>
> Human languages can be developed too. Doesn't langdev seem to allow
> for human language development?
>
Ugh... langdev makes me think of English.
devlang?
I'm not so good with human languages. I'm quite good with programming
languages. Are others the same?
It's unlikely you'll find solid agreement on a name. You initiated this
James. Pick something before you spend a year on it. Embrace your impulse.
Rod Pemberton
BGB
> "James Harris"<james.h...@googlemail.com> wrote in message
> news:21b10b52-3e4c-4ee8...@u3g2000vbj.googlegroups.com...
>>
>> Human languages can be developed too. Doesn't langdev seem to allow
>> for human language development?
>>
>
> Ugh... langdev makes me think of English.
>
> devlang?
>
> I'm not so good with human languages. I'm quite good with programming
> languages. Are others the same?
>
well, except that people who design natural languages are not as
commonly inclined towards terms like 'dev' or 'development', but more
often use terms like "conlang design" or "auxlang design", ...
> It's unlikely you'll find solid agreement on a name. You initiated this
> James. Pick something before you spend a year on it. Embrace your impulse.
>
well, one can make a career out of doing nothing, as a procrastinator.
in my case, it is funny, as some people accuse me of never really
settling on anything, and others accuse me of rushing into things with
no real plan. there is no win sometimes...
misc, ignore unless interested:
meanwhile, my own new-language-effort is going much more slowly than I
had hoped.
I got an assembler for the new bytecode format mostly written, but no
completed interpreter. this effort will take many months at this rate...
OTOH, I did add a few possibly useful features to my OO facilities and
x86 assembler.
OO facility tweaks:
field and method modifier flags are now supported (it also defines the
relevant modifier bits);
it is now possible to alias classes (basically, create class symlinks,
so that the same physical class can be accessed via multiple qnames);
...
x86 assembler additions: well, mostly a "resbto" special, which will
reserve bytes until a specific offset in the current section (mostly
useful for making executable structs).
also tweaked the metadata-database tool some, mostly so that it emits
modifier flag strings (strings are the canonical external
representation, rather than bit-masks). a relevant observation is also
that modifier flags tend to come in a small number of common
configurations (and very common in-fact that no modifiers are used).
meanwhile, for the flags I have recently had to resort to a 64-bit flags
field as I can't currently fit all the available modifiers into 32 bits.
as-is, a number of modifiers are already aliased ('sealed' and 'final'
use the same bit, as do 'native' and 'unsafe', ...). other aliased cases
depend on context, such as allowing a bit to be shared for cases where
modifier X only applies to methods and Y only applies to fields, ...
I am likely to modify the OO system such that fields and object
delegates share the same namespace (this will be the semantics in
BGBScript2, and nothing else currently uses object delegation in a way
which requires a separate namespace).
some other recent thoughts have involved giving the OO system the
ability to see C structs. this could be useful for BS2, and also
possibly to make a "slightly easier to use" interface for "safely"
reading/writing C structs to/from files (and could also ease
implementing a serialization interface which supports structs).
relevant modifiers:
"__transient", "__ltlendian" (little endian), "__bigendian", ... but
these will be wrapped in macros, probably: "TRANSIENT", "LTLENDIAN", and
"BIGENDIAN". that or I could just make them lower case, so it looks like
C just magically gained support for the "transient" keyword... (either
way, it will magically disappear if not being processed by my tools, for
sake of avoiding GCC or MSVC barfing when they see it...).
these would effect how structs are serialized, where the endianness
modifiers, well, indicate endianess, and transient indicates that a
given field should not be preserved (in the serialized data, it is
assumed not to exist).
I may consider an API call like:
int dyStructWriteBuffer(void *buf, void *obj, char *typename);
int dyStructReadBuffer(void *buf, void *obj, char *typename);
where: buf is the buffer, obj is the struct, and typename is the name of
whatever typedef is used (the existence of an exact-match typedef will
be mandatory).
int dyStructSizeof(char *typename);
int dyStructSizeofBuffer(char *typename);
would dynamically return:
the size of a struct (in memory), and the size of a struct when written
into a buffer (these may differ...). this interface would likely
prohibit non-transient pointer fields, ...
or such...
James Harris
...
> I was trying to focus you on "Terms of Service". For example, what
> happens to a registered (assuming one will be able to register, i.e.)
> participant's postings and such when they unregister? What about
> transient participant's posts and such? Will they be able to remove them?
> Modify them? Etc. The following kind of thing I find unacceptable (read,
> I won't register with a site that has any such TOS):
You mean "ownership of work"? As far as I'm concerned each person owns
his or her own work and could delete it at any time. If you are
worried about this be sure not to release your work to the public
domain as otherwise someone could copy it and add his own name,
couldn't he?
> "By posting content or making content available for inclusion on the
> <site name> forums, including without limitation threads and comments you
> post to the forum, you grant <site name> and its proprietors a
> world-wide, perpetual, irrevocable, royalty free, non-exclusive, fully
> sub-licensable license to use, distribute, reproduce, modify, adapt,
> publish, translate, publicly perform and publicly display, and to
> incorporate such content into other works in any format or medium now
> known or later developed."
I wouldn't find that acceptable either, far from it.
Would public domain allow all of that? A company - any company - could
copy public domain information and use it for any purpose.
Or perhaps the difference is that a company couldn't claim ownership
of something released to the public domain whereas the terms you
quoted seem to imply the site owners intend to become owners of
someone else's work.
Complex, isn't it!
James
James Harris
> > Human languages can be developed too. Doesn't langdev seem to allow
> > for human language development?
>
> Ugh... langdev makes me think of English.
>
> devlang?
>
> I'm not so good with human languages. I'm quite good with programming
> languages. Are others the same?
It depends what you mean by "good with."
> It's unlikely you'll find solid agreement on a name. You initiated this
> James. Pick something before you spend a year on it. Embrace your impulse.
It's not really about one person's choices. If we'd picked one already
we wouldn't have had the benefit of your comment, eh! The current pace
seems fine. I will post about hosting in the next few days. That
should be the last topic.
James
Tony
> On Jan 13, 11:51 am, "Tony" <nos...@myisp.net> wrote:
>
> ...
>
>> I was trying to focus you on "Terms of Service". For example, what
>> happens to a registered (assuming one will be able to register, i.e.)
>> participant's postings and such when they unregister? What about
>> transient participant's posts and such? Will they be able to remove
>> them? Modify them? Etc. The following kind of thing I find
>> unacceptable (read, I won't register with a site that has any such
>> TOS):
>
> You mean "ownership of work"?
"Nooooooo" (pronounced: "newwww"). Last time, you asked if I had read the
thread you referred me to. Well, lo and behold, I looked it up and
realized I had followed that thread (I mean, as much as my ADD would
allow ;)). Anyway, I still don't know what "you'all" are building.
Perhaps introductions of a sort are in order, cuz I don't know you or
your background, but I can't help but feel that I'm in a totally
different place than where you are coming from.
That said, you're a Del Fuego, right? (hehe, ;) ).
> As far as I'm concerned each person owns
> his or her own work and could delete it at any time.
So the implication is that you and cohorts are putting up a "SourceForge"
of sorts?
> If you are
> worried about this be sure not to release your work to the public
> domain as otherwise someone could copy it and add his own name,
> couldn't he?
Uh oh. K. I like text, and USENET, but it DOES have its limitations.
Metaphor time (oh noooo!). Yes, that. Something about little streams and
big rivers and water supply. (Mold your own metaphor! :).
>
>> "By posting content or making content available for inclusion on the
>> <site name> forums, including without limitation threads and
>> comments you post to the forum, you grant <site name> and its
>> proprietors a world-wide, perpetual, irrevocable, royalty free,
>> non-exclusive, fully sub-licensable license to use, distribute,
>> reproduce, modify, adapt, publish, translate, publicly perform and
>> publicly display, and to incorporate such content into other works
>> in any format or medium now known or later developed."
>
> I wouldn't find that acceptable either, far from it.
But how many of sites with similar TOSs do you subscribe to?
(rhetorical). Do you know? (rhetorical). Do you care? (rhetorical).
>
> Would public domain allow all of that? A company - any company - could
> copy public domain information and use it for any purpose.
Non-sequitor?
>
> Or perhaps the difference is
What, pray tell, are you comparing?
> that a company couldn't claim ownership
> of something released to the public domain
Explain this "fixation" with "public domain" please. (Or don't, it's not
important to me. I'm just typing back at you).
> whereas the terms you
> quoted seem to imply the site owners intend to become owners of
> someone else's work.
Oh. Ya think?
>
> Complex, isn't it!
Not to me. Is it for you? (rhetorical)
Tony
I'm just trying to figure out what you are trying to name! A dog named
Boo?
James Harris
> James Harris wrote:
> > On Jan 13, 11:51 am, "Tony" <nos...@myisp.net> wrote:
>
> > ...
>
> >> I was trying to focus you on "Terms of Service". For example, what
> >> happens to a registered (assuming one will be able to register, i.e.)
> >> participant's postings and such when they unregister? What about
> >> transient participant's posts and such? Will they be able to remove
> >> them? Modify them? Etc. The following kind of thing I find
> >> unacceptable (read, I won't register with a site that has any such
> >> TOS):
>
> > You mean "ownership of work"?
>
> "Nooooooo" (pronounced: "newwww"). Last time, you asked if I had read the
> thread you referred me to. Well, lo and behold, I looked it up and
> realized I had followed that thread (I mean, as much as my ADD would
> allow ;)). Anyway, I still don't know what "you'all" are building.
> Perhaps introductions of a sort are in order, cuz I don't know you or
> your background, but I can't help but feel that I'm in a totally
> different place than where you are coming from.
>
> That said, you're a Del Fuego, right? (hehe, ;) ).
>
> > As far as I'm concerned each person owns
> > his or her own work and could delete it at any time.
>
> So the implication is that you and cohorts are putting up a "SourceForge"
> of sorts?
No, not at all. Sourceforge appears to be a good place for sourceforge-
type stuff. Why set up something similar?
At least initially I'd use the proposed new site more for text than
programs. If I did put up some code it might be fixed code intended as
an example. If there was a demand for us to jointly *develop* programs
perhaps they could be hosted on Sourceforge and linked to from the
site. The web is, after all, a great medium for sharing and promoting
links to appropriate places.
I'm not saying we can store only text. There are probably many
legitimate uses for a common site that c.l.m'ers can share. For
example, say we wanted to develop a common IR. A shared web site would
be ideal for that.
...
> >> "By posting content or making content available for inclusion on the
> >> <site name> forums, including without limitation threads and
> >> comments you post to the forum, you grant <site name> and its
> >> proprietors a world-wide, perpetual, irrevocable, royalty free,
> >> non-exclusive, fully sub-licensable license to use, distribute,
> >> reproduce, modify, adapt, publish, translate, publicly perform and
> >> publicly display, and to incorporate such content into other works
> >> in any format or medium now known or later developed."
...
> > Or perhaps the difference is
>
> What, pray tell, are you comparing?
I think I was comparing the wording in the paragraph above and the
interpretation of ownership but let's not go there. It's not
important.
> > that a company couldn't claim ownership
> > of something released to the public domain
>
> Explain this "fixation" with "public domain" please. (Or don't, it's not
> important to me. I'm just typing back at you).
PD or the equivalent seemed the most appropriate licence for text
placed on a web site.
To reiterate, the site is not a forum but it would be a good place to
write the conclusions of a discussion.
James
Tony
But what?
> If I did put up some code it might be fixed code intended as
> an example.
"if you did"?
> If there was a demand for us
"us"? I think you need to let it go. I wasted a lot of TIME with that
kind of mindset. You are useless, but you have time. That is not to say
your parents are just in imposing taxes upon others to bore you. (That is
a waring action).
> to jointly *develop* programs
"and it'll rain money from L.D. Newsome and whiskey will flow from the
stream".Yep, you're a real real thinker there... god damn, Jack Twist,
you got it all figured out, don't ya."
"and
> perhaps they could be hosted on Sourceforge and linked to from the
> site. The web is, after all, a great medium for sharing and promoting
> links to appropriate places.
Sure, send your daughter to school on the interstate. I dare you.
>
> I'm not saying we
See ( I am not you or your "we" social group).
> can store only text. There are probably many
> legitimate uses for a common site that c.l.m'ers
there is no such thing.
> can share. For
> example, say we wanted to develop a common IR. A shared web site would
> be ideal for that.
Ref: "Easy Rider". Study it. Learn it.
James Harris
wrote:
OK. Votes were as follows
pldev for: 3 against: 1
langdev for: 3 against: 2
langdesign for: 2 against: 1
pldi for: 1 against: 1
parse this for: 1 against: 1
pldevel for: 0 against: 1
So pldev (for programming language development) it is.
I reserved www.pldev.tk a while ago for testing and Robbert has now
kindly reserved www.pldev.org (thanks, Robbert). www.pldev.org is
intended to be the final name of the site. Discussions about hosting
are in a different thread so see that thread for such discussions.
Sorry it was not possible to satisfy the suggestions of all but thanks
to everyone for the input and opinions.
James
aury
wrote:> > Even if it is just a club for an existing group of people or for
> > hobbyists though, be aware that the ownership of content (which looks
> > like it will come from participants) and the terms of agreement of the
> > usage of such content now and in perpetuity is very, very important to
> > have documented clearly, even if such terms are implicit from governing
> > laws, and that they be understood by all regular participants and
> > transients.
>
> Tony has raised a good point. It is important to know what licence
> website content will be published under. Any suggestions?
>
> Do we need to distinguish the terms for written articles from the
> terms for source code? And maybe other types?
>
> AIUI, regardless of licensing terms, ownership and *copyright* is
> generally retained by the author (unless employed to write in which
> case copyright may be held by the employer). I think of licensing as
> relating to what people can do with the work.
>
> James
Idea for new site about creating new programming languages
are fine.
And why must be C or C++ ?
Robbert Haarman
I have read the posts about the licensing for the content of the site
about programming languages, and have some thoughts of my own to share.
First, it is clear that licensing is a hot topic, so I am probably going
to step on a few people's toes at some point. Apologies in advance.
Secondly, I am not a copyright lawyer, and certainly not an expert on how
copyright works everywhere around the world, so my ideas of how things
work from a legal perspective may be completely off. With that out of
the way, here are my actual ideas:
It seems to me that it would be best to look at what some successful
user-contributed content sites do, and do the same. Not only because
that evidently works, but also because this makes life easier for would-be
contributors - and the less of an effort we require from contributors,
the better.
So what examples do we have? I think _the_ example of a successful
user-contributed content website is Wikipedia. Another example would be
Slashdot, which may not have as much mindshare, but has been around
for a very long time. And, of course, there is Usenet, which has been
around even longer, and which we all post to - although I am not sure
the copyright/licensing issues are clearly defined here.
What is clear is that there should be some understanding that content
contributed by users will be displayed on the site. As the Slashdot
terms of use state:
> With respect to text or data entered into and stored by
> publicly-accessible site features such as forums, comments and bug
> trackers ("Geeknet Public Content"), the submitting user retains
> ownership of such Geeknet Public Content;
and
> In each such case, the submitting user grants Geeknet the
> royalty-free, perpetual, irrevocable, non-exclusive, transferable
> license to use, reproduce, modify, adapt, publish, translate, create
> derivative works from, distribute, perform, and display such Content
> (in whole or part) worldwide and/or to incorporate it in other works
> in any form, media, or technology now known or later developed, all
> subject to the terms of any applicable license.
In other words, user-contributed content stays owned by the user, but
Geeknet gets to do what they want with it - including transfering those
rights to others.
Wikipedia's terms of use regarding licensing are a bit too large to
quote here, but they can be found at
http://wikimediafoundation.org/wiki/Terms_of_Use What it boils down to
is that if you contribute text, you license that text to Wikimedia under
the Creative Commons Attribution-ShareAlike 3.0 Unported License,
and, because that's what they originally started out with, also the
GNU Free Documentation License.
On Usenet, as I said, I don't think the copyright/licensing issues are
completely clear. If I had to guess, I would say that, by posting to
Usenet, you implicitly grant permission to publish your message, but
you remain the copyright holder and otherwise retain all rights.
Any of these approaches seems fine to me, and, given the number of
contributors to each of these services, evidently a lot of other people
are fine with them, too. My vote is for keeping it simple - the site
gets the right to publish your contribution in whole and in part, and
otherwise the contributor stays the owner and retains responsibility.
Wikipedia's terms are a bit complex because of the two licenses, but
if we copy their model, we could reduce that to just a Creative Commons
license.
I don't recommend using a software license, because I imagine our
content will mostly be text - with source code which could likely be
used under fair use. If the need arises, we can always introduce a
license specifically for the program code later. However, I want to
avoid complicating things by having multiple licenses.
Regards,
Bob
Tony
"Robbert Haarman" <comp.la...@inglorion.net> wrote in message
news:2011020514...@yoda.inglorion.net...
Two things: First, people (er, sheeple) regularly "sign their rights
away", as they have been conditioned to do so. Being "fine with them" may
just be naivete. Second, who is to say what the content quality would
have been (higher) had it not been withheld by the intellectual property
holder because of terms that were unacceptable.
The above is not to say that just having the right TOA will garner
higher-quality content, but is to suggest that having a TOA that is
unacceptable will indeed prevent some people from posting certain
material or even posting at all. (There is something fundamental hinted
at in this statement but I don't want to "go there" at this point).
(Aside: Remember, Wikipedia doesn't push any boundary as far as creating
or synthesizing. It only documents what is "common" knowledge. That is
obviously vastly different from a site where persons discuss new things,
create, synthesize, etc.)
Tony
> OK. Votes were as follows
>
> pldev for: 3 against: 1
> langdev for: 3 against: 2
> langdesign for: 2 against: 1
> pldi for: 1 against: 1
> parse this for: 1 against: 1
> pldevel for: 0 against: 1
>
> So pldev (for programming language development) it is.
>
What's interesting in the above is how small the group is.
Rod Pemberton
news:6nj4p.50781$3K....@newsfe12.iad...
Yeah, when I saw it, I thought James couldn't count... :-0
He clearly didn't count "for" vote for each name by those that proposed the
name. E.g., if he had, "parse this" would've had at least two "for" votes,
mine for suggesting it, and someone in this thread who said they liked it.
AISI, the "against" and "for" for it must've been by for those who voted by
email only... It's not a big deal. "pldev" and "langdev" are tied for the
win. It's still up to James to pick the name... ;-) I don't think you're
getting out of it James! rofl.
Rod Pemberton
James Harris
wrote:> > James Harris wrote:
...
> > > OK. Votes were as follows
>
> > > pldev for: 3 against: 1
> > > langdev for: 3 against: 2
> > > langdesign for: 2 against: 1
> > > pldi for: 1 against: 1
> > > parse this for: 1 against: 1
> > > pldevel for: 0 against: 1
>
> > > So pldev (for programming language development) it is.
>
> > What's interesting in the above is how small the group is.
>
> Yeah, when I saw it, I thought James couldn't count... :-0
>
> He clearly didn't count "for" vote for each name by those that proposed the
> name. E.g., if he had, "parse this" would've had at least two "for" votes,
> mine for suggesting it, and someone in this thread who said they liked it.
> AISI, the "against" and "for" for it must've been by for those who voted by
> email only... It's not a big deal. "pldev" and "langdev" are tied for the
> win. It's still up to James to pick the name... ;-) I don't think you're
> getting out of it James! rofl.
:-)
I didn't include the initial suggestions as votes as they, including
those I put forward, were just suggestions. At least that's how I took
them. For example, I suggested pld but didn't put in a vote for it so
it scored zero (or, with a negative, minus 1). What I counted were the
votes that followed on from the following post containing the list of
suggestions.
http://groups.google.com/group/comp.lang.misc/msg/053f660ab4859656?hl=en
I could have included initial suggestions as votes but it didn't seem
appropriate and, I think, would have muddied the waters. For example,
if person P thought of a good name but saw that someone else had
already suggested it P wouldn't be likely to "suggest" it again
leading to an artificially low vote for a good name.
I did think about ranking votes but it would have meant asking someone
who had already expressed a liking for more than one name whether his
or her preferences should be weighted and I guessed that would be
straying too far into Usenet fatigue territory *and* it raised issues
over how to weight others' comments. So I went for the simple option:
each positive comment counted one for and each negative comment
counted as one against.
It's impossible to please all the people all the time, eh. :-(
James
James Harris
wrote:
...
> What is clear is that there should be some understanding that content
> contributed by users will be displayed on the site. As the Slashdot
> terms of use state:
>
> > With respect to text or data entered into and stored by
> > publicly-accessible site features such as forums, comments and bug
> > trackers ("Geeknet Public Content"), the submitting user retains
> > ownership of such Geeknet Public Content;
>
> and
>
> > In each such case, the submitting user grants Geeknet the
> > royalty-free, perpetual, irrevocable, non-exclusive, transferable
> > license to use, reproduce, modify, adapt, publish, translate, create
> > derivative works from, distribute, perform, and display such Content
> > (in whole or part) worldwide and/or to incorporate it in other works
> > in any form, media, or technology now known or later developed, all
> > subject to the terms of any applicable license.
>
> In other words, user-contributed content stays owned by the user, but
> Geeknet gets to do what they want with it - including transfering those
> rights to others.
I'm a bit puzzled by the niceties of the distinction in the above
paragraph. The user stays the owner but Geeknet gets to do what it
wants with the content?
Is it practical to write a custom licence? I imagined that needed
professional legal knowledge. On the other hand I like the idea. I'd
like to include comments on the purpose being for information only, it
coming with no guarantees and disclaiming all liability, as far as is
possible.
Some sites do seem to have their own Ts&Cs. For example, I recently
requested an account on Heliohost and their terms are not at all
legalese:
http://www.heliohost.org/h-terms.html
The terms are short but they don't seem watertight.
Another thought (this is a murky business, isn't it) is that with a
bespoke licence and maybe even with a standard licence the licensing
of the content and the terms applicable to anyone who edits will
almost certainly need to change, i.e. be improved, after content is
already present. Unless we get these perfect first time that seems
like there needs to be a clause to allow changes to the terms.
For example, not that I think we will want to change to more onerous
terms but in principle, if the terms change then people who have
already contributed content may want to remove what they have added.
I think I need to go and lie down in a darkened room. :-(
Bottom line: I like the idea of bespoke terms. Can we successfully
define them?
James
aury
So,what would be on the end of this discussion?
Do you finaly find some place ?
James Harris
...
> So,what would be on the end of this discussion?
> Do you finaly find some place ?
Please take a look at
http://groups.google.com/group/comp.lang.misc/msg/d1fcf571d94fb830?hl=en
None of the three test sites are ideal and we are looking for a better
host. Sorry it's taking so long.
James