Eclipse Community Forums
Forum Search:

Search      Help    Register    Login    Home
Home » Archived » IMP » IMP release time
IMP release time [message #2261] Fri, 24 August 2007 02:47 Go to next message
Auerliano is currently offline AuerlianoFriend
Messages: 149
Registered: July 2009
Senior Member
Bob,

In ibm safari website, it says if you want to download safari 'click
here', and then, when I 'clicked there' it said send a message to the
researcher, which I think it wass you. So I sent a request for trying
safari, but never heard back.

Now I found this imp thing, which I think the forced renaming of it was
a good idea. It's a pain when you want to search for technical help of
some piece of software with a very common name and spelling, and even
worse, when it has the same name as some other widely-used software.
(Even 'eclipse' was not a good idea, I feel sorry for the astromoner
guys as ibm eclipse has invaded their 'google space'.)

(1) Is it possible to evaluate safari before it is turned to imp?

(2) Is imp going to be available in october 2007, or there would be a delay?

(3) Would the initial release of imp provide a mature working safari, or
there is a long way to meet the features given in the presentation of
eclipse-07?
Re: IMP release time [message #2291 is a reply to message #2261] Sat, 25 August 2007 21:25 Go to previous messageGo to next message
Robert M. Fuhrer is currently offline Robert M. FuhrerFriend
Messages: 294
Registered: July 2009
Senior Member
Ali wrote:
> Bob,
>
> In ibm safari website, it says if you want to download safari 'click
> here', and then, when I 'clicked there' it said send a message to the
> researcher, which I think it wass you. So I sent a request for trying
> safari, but never heard back.

Sorry not to have gotten back to you; we got so many such replies that
we just couldn't keep up. We had to be selective initially so as not to
get overwhelmed for support while the software was still in an early
state. It's still not 100% polished, of course, but it's gotten quite
a bit better.

> Now I found this imp thing, which I think the forced renaming of it was
> a good idea. It's a pain when you want to search for technical help of
> some piece of software with a very common name and spelling, and even
> worse, when it has the same name as some other widely-used software.
> (Even 'eclipse' was not a good idea, I feel sorry for the astromoner
> guys as ibm eclipse has invaded their 'google space'.)
>
> (1) Is it possible to evaluate safari before it is turned to imp?

Good news: We just sent a snapshot of our CVS repository to the
sourceforge admin folks, and hopefully it should actually become
visible to the public within a couple of days or so.

We're in the process of getting the web materials and bugzilla
site in place as well. It should all happen within a matter of
days at this point.

> (2) Is imp going to be available in october 2007, or there would be a
> delay?

If you're referring to the eclipse.org project, that depends entirely
on the eclipse community. If enough people voice enough support for the
project, then it will happen; otherwise, we'll still have the sourceforge
project.

> (3) Would the initial release of imp provide a mature working safari, or
> there is a long way to meet the features given in the presentation of
> eclipse-07?

Depends on what you think "mature" means. We did try to make it clear in
the EclipseCon '07 presentation what was already implemented and what
was planned, but I imagine there could still be some questions.

IMP should now be usable, and supports a nontrivial set of language/IDE
services. We have more or less reasonable support for the things we
demoed at EclipseCon: syntax highlighting, source folding, hyperlinking,
hover help, outlining, project building, and a few other IDE services.

At this point, the things IMP doesn't support well (or at all) include
indexed search, "language embedding" (e.g. assembler in C, or SQL in Java),
source formatting, and launching & debugging. We're just starting to work
on language embedding, and will start work on formatting soon (with some
colleagues at CWI). We've got ideas for improving error handling (at the
parsing and higher levels), and hopefully will get to work on that in the
near future. We're also whittling away at our type analysis engine (e.g.
in support of refactoring, based on our work on the Eclipse type-related
JDT refactorings in 3.1 and 3.2). IMP doesn't have adapters to frameworks
like JastAdd or ANTLR, and this is an area where the community could help
a lot in fleshing things out. IMP has nothing for quick fixes at present.

The above isn't meant to be an exhaustive list of what IMP does/doesn't
handle, but should give you a basic idea of where things stand.

Part of the point of releasing IMP to the open-source community is to get
people involved in helping to refine the vision and the implementation so
that it meets people's needs better than one team is able to anticipate,
and to encourage others to contribute components and ideas that we don't
have the bandwidth to supply. In other words, we're opening things up not
because everything is already 100% polished, but because no single team
can really expect to make the vision a reality. Plus, we shouldn't have
*all* the fun! :-)

Anyway, hopefully this is just the beginning of the discussion!

Again, sorry you had to wait to get your hands on the software and join
the effort, but please feel free to contribute with whatever resources
and ideas you have. We're really excited about the potential for Eclipse
in particular, as well as for language tool development in general!

Cheers,
- Bob
Re: IMP release time [message #2321 is a reply to message #2291] Sun, 26 August 2007 23:31 Go to previous messageGo to next message
Auerliano is currently offline AuerlianoFriend
Messages: 149
Registered: July 2009
Senior Member
Bob,

Thanks a lot for the reply -- it was illuminating. Could you provide the
sf.net address for the project?

I have posted a message to technology.dltk asking the dltk folk about
the difference between the two projects. It would be good if imp and
dltk could be somehow merged, it seems to me the two projects are -more
or less-in parallel. Moreover, it seems that there is this potential to
have a united project where the community can benefit a great
ide-generator with a double-speed progress.


Robert M. Fuhrer wrote:
> Ali wrote:
>> Bob,
>>
>> In ibm safari website, it says if you want to download safari 'click
>> here', and then, when I 'clicked there' it said send a message to the
>> researcher, which I think it wass you. So I sent a request for trying
>> safari, but never heard back.
>
> Sorry not to have gotten back to you; we got so many such replies that
> we just couldn't keep up. We had to be selective initially so as not to
> get overwhelmed for support while the software was still in an early
> state. It's still not 100% polished, of course, but it's gotten quite
> a bit better.
>
>> Now I found this imp thing, which I think the forced renaming of it
>> was a good idea. It's a pain when you want to search for technical
>> help of some piece of software with a very common name and spelling,
>> and even worse, when it has the same name as some other widely-used
>> software. (Even 'eclipse' was not a good idea, I feel sorry for the
>> astromoner guys as ibm eclipse has invaded their 'google space'.)
>>
>> (1) Is it possible to evaluate safari before it is turned to imp?
>
> Good news: We just sent a snapshot of our CVS repository to the
> sourceforge admin folks, and hopefully it should actually become
> visible to the public within a couple of days or so.
>
> We're in the process of getting the web materials and bugzilla
> site in place as well. It should all happen within a matter of
> days at this point.
>
>> (2) Is imp going to be available in october 2007, or there would be a
>> delay?
>
> If you're referring to the eclipse.org project, that depends entirely
> on the eclipse community. If enough people voice enough support for the
> project, then it will happen; otherwise, we'll still have the sourceforge
> project.
>
>> (3) Would the initial release of imp provide a mature working safari,
>> or there is a long way to meet the features given in the presentation
>> of eclipse-07?
>
> Depends on what you think "mature" means. We did try to make it clear in
> the EclipseCon '07 presentation what was already implemented and what
> was planned, but I imagine there could still be some questions.
>
> IMP should now be usable, and supports a nontrivial set of language/IDE
> services. We have more or less reasonable support for the things we
> demoed at EclipseCon: syntax highlighting, source folding, hyperlinking,
> hover help, outlining, project building, and a few other IDE services.
>
> At this point, the things IMP doesn't support well (or at all) include
> indexed search, "language embedding" (e.g. assembler in C, or SQL in Java),
> source formatting, and launching & debugging. We're just starting to work
> on language embedding, and will start work on formatting soon (with some
> colleagues at CWI). We've got ideas for improving error handling (at the
> parsing and higher levels), and hopefully will get to work on that in the
> near future. We're also whittling away at our type analysis engine (e.g.
> in support of refactoring, based on our work on the Eclipse type-related
> JDT refactorings in 3.1 and 3.2). IMP doesn't have adapters to frameworks
> like JastAdd or ANTLR, and this is an area where the community could help
> a lot in fleshing things out. IMP has nothing for quick fixes at present.
>
> The above isn't meant to be an exhaustive list of what IMP does/doesn't
> handle, but should give you a basic idea of where things stand.
>
> Part of the point of releasing IMP to the open-source community is to get
> people involved in helping to refine the vision and the implementation so
> that it meets people's needs better than one team is able to anticipate,
> and to encourage others to contribute components and ideas that we don't
> have the bandwidth to supply. In other words, we're opening things up not
> because everything is already 100% polished, but because no single team
> can really expect to make the vision a reality. Plus, we shouldn't have
> *all* the fun! :-)
>
> Anyway, hopefully this is just the beginning of the discussion!
>
> Again, sorry you had to wait to get your hands on the software and join
> the effort, but please feel free to contribute with whatever resources
> and ideas you have. We're really excited about the potential for Eclipse
> in particular, as well as for language tool development in general!
>
> Cheers,
> - Bob
DTLK & IMP [message #2351 is a reply to message #2321] Mon, 27 August 2007 12:54 Go to previous message
Eclipse UserFriend
Originally posted by: brunodomedeiros+spam.com.gmail

Ali wrote:
> Bob,
>
> Thanks a lot for the reply -- it was illuminating. Could you provide the
> sf.net address for the project?
>
> I have posted a message to technology.dltk asking the dltk folk about
> the difference between the two projects. It would be good if imp and
> dltk could be somehow merged, it seems to me the two projects are -more
> or less-in parallel. Moreover, it seems that there is this potential to
> have a united project where the community can benefit a great
> ide-generator with a double-speed progress.
>
>

Regarding the DTLK & IMP issue, I also want to repost some comments I
made in the DTLK NG: It seems to me there is a lot of overlap between
the two projects. Sure, it is not a complete overlap, there may be many
things DLTK aims to provide that IMP doesn't, and vice-versa, but there
is a great body of work that is common, mostly JDT infrastructure.

For example the IMP proposal states that "The DLTK is a related effort,
but requires the use of a specific AST class hierarchy to drive much of
its functionality, which is not suitable for arbitrary languages, nor
for situations in which a compiler front-end already exists.", but that
is not the case: I have started integrating with DLTK a previous Eclipse
IDE for the D programming language (which is static and natively
compiled, btw) and even tough DLTK's AST hierarchy wasn't used (other
than the ASTNode base class) I've found that there was still a lot to be
gained with the integration. (my thread in the DLTK NG details it further)

So, since IMP is in the proposal phase, and looking for comments, I hope
you take this into consideration, and hopefully the two project teams
could talk some more to give more thought to this issue.

Thanks.

--
Bruno Medeiros - MSc in CS/E student
Re: IMP release time [message #565679 is a reply to message #2261] Sat, 25 August 2007 21:25 Go to previous message
Robert M. Fuhrer is currently offline Robert M. FuhrerFriend
Messages: 294
Registered: July 2009
Senior Member
Ali wrote:
> Bob,
>
> In ibm safari website, it says if you want to download safari 'click
> here', and then, when I 'clicked there' it said send a message to the
> researcher, which I think it wass you. So I sent a request for trying
> safari, but never heard back.

Sorry not to have gotten back to you; we got so many such replies that
we just couldn't keep up. We had to be selective initially so as not to
get overwhelmed for support while the software was still in an early
state. It's still not 100% polished, of course, but it's gotten quite
a bit better.

> Now I found this imp thing, which I think the forced renaming of it was
> a good idea. It's a pain when you want to search for technical help of
> some piece of software with a very common name and spelling, and even
> worse, when it has the same name as some other widely-used software.
> (Even 'eclipse' was not a good idea, I feel sorry for the astromoner
> guys as ibm eclipse has invaded their 'google space'.)
>
> (1) Is it possible to evaluate safari before it is turned to imp?

Good news: We just sent a snapshot of our CVS repository to the
sourceforge admin folks, and hopefully it should actually become
visible to the public within a couple of days or so.

We're in the process of getting the web materials and bugzilla
site in place as well. It should all happen within a matter of
days at this point.

> (2) Is imp going to be available in october 2007, or there would be a
> delay?

If you're referring to the eclipse.org project, that depends entirely
on the eclipse community. If enough people voice enough support for the
project, then it will happen; otherwise, we'll still have the sourceforge
project.

> (3) Would the initial release of imp provide a mature working safari, or
> there is a long way to meet the features given in the presentation of
> eclipse-07?

Depends on what you think "mature" means. We did try to make it clear in
the EclipseCon '07 presentation what was already implemented and what
was planned, but I imagine there could still be some questions.

IMP should now be usable, and supports a nontrivial set of language/IDE
services. We have more or less reasonable support for the things we
demoed at EclipseCon: syntax highlighting, source folding, hyperlinking,
hover help, outlining, project building, and a few other IDE services.

At this point, the things IMP doesn't support well (or at all) include
indexed search, "language embedding" (e.g. assembler in C, or SQL in Java),
source formatting, and launching & debugging. We're just starting to work
on language embedding, and will start work on formatting soon (with some
colleagues at CWI). We've got ideas for improving error handling (at the
parsing and higher levels), and hopefully will get to work on that in the
near future. We're also whittling away at our type analysis engine (e.g.
in support of refactoring, based on our work on the Eclipse type-related
JDT refactorings in 3.1 and 3.2). IMP doesn't have adapters to frameworks
like JastAdd or ANTLR, and this is an area where the community could help
a lot in fleshing things out. IMP has nothing for quick fixes at present.

The above isn't meant to be an exhaustive list of what IMP does/doesn't
handle, but should give you a basic idea of where things stand.

Part of the point of releasing IMP to the open-source community is to get
people involved in helping to refine the vision and the implementation so
that it meets people's needs better than one team is able to anticipate,
and to encourage others to contribute components and ideas that we don't
have the bandwidth to supply. In other words, we're opening things up not
because everything is already 100% polished, but because no single team
can really expect to make the vision a reality. Plus, we shouldn't have
*all* the fun! :-)

Anyway, hopefully this is just the beginning of the discussion!

Again, sorry you had to wait to get your hands on the software and join
the effort, but please feel free to contribute with whatever resources
and ideas you have. We're really excited about the potential for Eclipse
in particular, as well as for language tool development in general!

Cheers,
- Bob
Re: IMP release time [message #565695 is a reply to message #2291] Sun, 26 August 2007 23:31 Go to previous message
Auerliano is currently offline AuerlianoFriend
Messages: 149
Registered: July 2009
Senior Member
Bob,

Thanks a lot for the reply -- it was illuminating. Could you provide the
sf.net address for the project?

I have posted a message to technology.dltk asking the dltk folk about
the difference between the two projects. It would be good if imp and
dltk could be somehow merged, it seems to me the two projects are -more
or less-in parallel. Moreover, it seems that there is this potential to
have a united project where the community can benefit a great
ide-generator with a double-speed progress.


Robert M. Fuhrer wrote:
> Ali wrote:
>> Bob,
>>
>> In ibm safari website, it says if you want to download safari 'click
>> here', and then, when I 'clicked there' it said send a message to the
>> researcher, which I think it wass you. So I sent a request for trying
>> safari, but never heard back.
>
> Sorry not to have gotten back to you; we got so many such replies that
> we just couldn't keep up. We had to be selective initially so as not to
> get overwhelmed for support while the software was still in an early
> state. It's still not 100% polished, of course, but it's gotten quite
> a bit better.
>
>> Now I found this imp thing, which I think the forced renaming of it
>> was a good idea. It's a pain when you want to search for technical
>> help of some piece of software with a very common name and spelling,
>> and even worse, when it has the same name as some other widely-used
>> software. (Even 'eclipse' was not a good idea, I feel sorry for the
>> astromoner guys as ibm eclipse has invaded their 'google space'.)
>>
>> (1) Is it possible to evaluate safari before it is turned to imp?
>
> Good news: We just sent a snapshot of our CVS repository to the
> sourceforge admin folks, and hopefully it should actually become
> visible to the public within a couple of days or so.
>
> We're in the process of getting the web materials and bugzilla
> site in place as well. It should all happen within a matter of
> days at this point.
>
>> (2) Is imp going to be available in october 2007, or there would be a
>> delay?
>
> If you're referring to the eclipse.org project, that depends entirely
> on the eclipse community. If enough people voice enough support for the
> project, then it will happen; otherwise, we'll still have the sourceforge
> project.
>
>> (3) Would the initial release of imp provide a mature working safari,
>> or there is a long way to meet the features given in the presentation
>> of eclipse-07?
>
> Depends on what you think "mature" means. We did try to make it clear in
> the EclipseCon '07 presentation what was already implemented and what
> was planned, but I imagine there could still be some questions.
>
> IMP should now be usable, and supports a nontrivial set of language/IDE
> services. We have more or less reasonable support for the things we
> demoed at EclipseCon: syntax highlighting, source folding, hyperlinking,
> hover help, outlining, project building, and a few other IDE services.
>
> At this point, the things IMP doesn't support well (or at all) include
> indexed search, "language embedding" (e.g. assembler in C, or SQL in Java),
> source formatting, and launching & debugging. We're just starting to work
> on language embedding, and will start work on formatting soon (with some
> colleagues at CWI). We've got ideas for improving error handling (at the
> parsing and higher levels), and hopefully will get to work on that in the
> near future. We're also whittling away at our type analysis engine (e.g.
> in support of refactoring, based on our work on the Eclipse type-related
> JDT refactorings in 3.1 and 3.2). IMP doesn't have adapters to frameworks
> like JastAdd or ANTLR, and this is an area where the community could help
> a lot in fleshing things out. IMP has nothing for quick fixes at present.
>
> The above isn't meant to be an exhaustive list of what IMP does/doesn't
> handle, but should give you a basic idea of where things stand.
>
> Part of the point of releasing IMP to the open-source community is to get
> people involved in helping to refine the vision and the implementation so
> that it meets people's needs better than one team is able to anticipate,
> and to encourage others to contribute components and ideas that we don't
> have the bandwidth to supply. In other words, we're opening things up not
> because everything is already 100% polished, but because no single team
> can really expect to make the vision a reality. Plus, we shouldn't have
> *all* the fun! :-)
>
> Anyway, hopefully this is just the beginning of the discussion!
>
> Again, sorry you had to wait to get your hands on the software and join
> the effort, but please feel free to contribute with whatever resources
> and ideas you have. We're really excited about the potential for Eclipse
> in particular, as well as for language tool development in general!
>
> Cheers,
> - Bob
DTLK & IMP [message #565721 is a reply to message #2321] Mon, 27 August 2007 12:54 Go to previous message
Bruno Medeiros is currently offline Bruno MedeirosFriend
Messages: 14
Registered: July 2009
Junior Member
Ali wrote:
> Bob,
>
> Thanks a lot for the reply -- it was illuminating. Could you provide the
> sf.net address for the project?
>
> I have posted a message to technology.dltk asking the dltk folk about
> the difference between the two projects. It would be good if imp and
> dltk could be somehow merged, it seems to me the two projects are -more
> or less-in parallel. Moreover, it seems that there is this potential to
> have a united project where the community can benefit a great
> ide-generator with a double-speed progress.
>
>

Regarding the DTLK & IMP issue, I also want to repost some comments I
made in the DTLK NG: It seems to me there is a lot of overlap between
the two projects. Sure, it is not a complete overlap, there may be many
things DLTK aims to provide that IMP doesn't, and vice-versa, but there
is a great body of work that is common, mostly JDT infrastructure.

For example the IMP proposal states that "The DLTK is a related effort,
but requires the use of a specific AST class hierarchy to drive much of
its functionality, which is not suitable for arbitrary languages, nor
for situations in which a compiler front-end already exists.", but that
is not the case: I have started integrating with DLTK a previous Eclipse
IDE for the D programming language (which is static and natively
compiled, btw) and even tough DLTK's AST hierarchy wasn't used (other
than the ASTNode base class) I've found that there was still a lot to be
gained with the integration. (my thread in the DLTK NG details it further)

So, since IMP is in the proposal phase, and looking for comments, I hope
you take this into consideration, and hopefully the two project teams
could talk some more to give more thought to this issue.

Thanks.

--
Bruno Medeiros - MSc in CS/E student
Previous Topic:IMP release time
Next Topic:Unit testing support
Goto Forum:
  


Current Time: Mon Dec 22 18:40:15 GMT 2014

Powered by FUDForum. Page generated in 0.01701 seconds
.:: Contact :: Home ::.

Powered by: FUDforum 3.0.2.
Copyright ©2001-2010 FUDforum Bulletin Board Software