An interesting project using MPS to do augmented embedded programming

[Jake Brownson]

First of all has anybody used MPS (http://www.jetbrains.com/mps/)? I
spent over a year with it and ran into several extremely frustrating
problems that eventually made me ditch it and take a new approach with
my project, but a former colleague of mine has been pretty successful
with it.

His project is mbeddr: http://mbeddr.com

He has a great intro video on the site. It's basically a structured
editor for a sane subset of C with some interesting modular extensions
and some nice projections.

It's even open source if you want to dig into it.

Thought I'd share it w/ the group.

[David Barbour]

Your colleague has done some impressive work.

Hmm. Now I'm tempted to add unit information to numbers in my code. I'd forgotten how useful those can be.

--
You received this message because you are subscribed to the Google Groups "Augmented Programming" group.
To unsubscribe from this group and stop receiving emails from it, send an email to augmented-progra...@googlegroups.com.
To post to this group, send email to augmented-...@googlegroups.com.
Visit this group at http://groups.google.com/group/augmented-programming.
For more options, visit https://groups.google.com/groups/opt_out.

[Boaz Rosenan]

I was introduced to mbeddr back in 2010, in a SPLASH tutorial by Markus Voelter, before it even had a name.  Impressive indeed.

As for MPS, I owe it the most significant breakthrough in my research, as it made me understand what projectional editing is, and how it can be used for a programming language.  That was back in 2005.  After it was officially released in 2009, I did some work comparing it to my own projectional editing solution (the Cedalion programming language).

My impression of it is mixed.  On the one hand, one can notice the amount of investment made in making projectional editing as smooth as it can be, with some use-cases being almost as fluent as writing Java code in e.g., Eclipse.  On the other hand (and this is where I can relate to your frustration), I felt it was too rigid at times.  I don't remember a concrete example, but I think it would not allow me to use things that are not (yet) properly defined.  I remember some "chicken and egg" problems, especially when editing Java code.  It would get into a point where I needed to think really hard how to work around MPS...

I think it is a good thing that there is something such as MPS (and that it is mostly opensource), but I wouldn't use it for my next software project.  Actually, I've known MPS for 8 years now, and repeatedly chose not to use it for any the software projects I started since.

P.S., I think Markus Voelter, the original author of mbeddr, is the best thing that ever happened to MPS (is he the colleague you mentioned?). Aside from mbeddr, he developed a tool that allows MPS code to be version-controlled over GIT, and takes every opportunity to give talks about MPS and the cool things he has done with it.  If you look for academic papers about MPS you will find mostly ones authored (or co-authored) by Voelter.

[Kyle Murray (krilnon)]

Which came first, JetBrains' MPS or Simonyi / Intentional Software's editor?  I've always thought that they were pretty similar and that Intentional did it first, but I could have my timelines confused.

[Boaz Rosenan]

I believe Simonyi started working on Intentional back in the 1990's, when at Microsoft Research.  They made some publications (at Microsoft first, than at Simonyi's own company Intentional Software) starting 1996 and onward.  I saw a video presentation they gave in 2009, which I believe to be the first public appearance the actual tool made.

I do not know when actual work on MPS started, but the first mention of it I encountered was in 2004, in an article by Sergei Dmitriev titled "Language Oriented Programming".  I think the software was first made public in 2005, with an "early access program", and finally was released in 2009.

So the short answer, is that work on Intentional probably started first, but MPS was probably released earlier.

--

[Sean McDirmid]

I saw Simonyi present the work at UW in 1997, I think. There was even some kind of demo.

Sent from my Windows Phone

---

From: Boaz Rosenan
Sent: ‎11/‎12/‎2013 17:18
To: augmented-...@googlegroups.com
Subject: Re: An interesting project using MPS to do augmented embedded programming