On Thu, Jul 7, 2011 at 2:26 PM, Offray Vladimir Luna Cárdenas <[email protected]> wrote:
> This is kind of strange synchrony (again) with my inquiries and the Leo > word. I have been away of this list for a while when I was in my > immersion in the Smalltalk/Pharo[1] world when I found [2]Moose software > analysis tool and it kind of remind me Leo. My plan now is to use > Pharo/Moose and Python/Leo pair to pair to see some cross-pollination > between the ideas of both. > > [1] http://www.pharo-project.org/ > [2] http://www.moosetechnology.org/ Many thanks for these links. On my walk yesterday I had many new thoughts. One was that Leo's leaders, including you, are likely to come up with the way forward. That's been true many times in the past: I had a vague idea and somebody else showed me how to do it :-) Putting the "analyst in the center" might be a good summary of the present state of the "lint" work. Type inference, of whatever kind, is only one type of analysis that I would like to do on Leo. It's good to know that other projects are thinking along the same lines... > For me Smalltalk word has tried to build a comprehensive discourse about > informatics, one that, in words of Alan Kay, is trying to go from > "cooper to the user", from virtual machines, to user interfaces. In this > intend they develop a self described deeply interactive system in a > uniform objectual discourse. Now with tools like Moose, they're using > this environment to bring light about other software constructs of > informatics. For me Leo is an alternative approach for the same problem. It seems like it. > I'm not a programmer, but the use of outlines and clones to organize my > information is this kind of proto-discourse about my interaction with > the computer in tree form. Now I plan to "live inside Leo" for most of > my projects to see how much of this discourse can be build and how Leo > can help me to bootstrap understanding in dealing with complexity of > heterogeneous computer systems. Please keep us informed about your thoughts and experiences. > Sorry if, for the moment, this sound > kind of abstract, but this is the kind of thinking that Leo/Smalltalk > provokes and invites in me, and I will give more concrete details about > the projects after. No need for apology. This is just about exactly the kind of thinking I use when in creative mode. I'm not sure I would call it abstract exactly. I might call it vague :-) Everything is vague when we don't know exactly where we are going. However, we do know, vaguely or not, that we are only just at the beginning of what can be done with the analysis of programs. That's the important thing. >> 2. Thinking about lint and program analysis leads me to think of Leo >> as a platform for new kinds of programming tools. > > I'm agree. Caliopy is an example of this kind of new programming tools, > an also I think that the 5.0 with the attempt to bring more users needs > to "think outside the box" of programmers and think in a more general > user (something like caliopy for structural engineers but thinking in a > wider audience). Interesting. Caliopy's buttons show just how much can be done already in the Leo framework. BTW, googling Leo and Caliopy I found some files at the University of Namibia. One never knows where the ideas will spread. Speaking of which, I basically have no idea about how many people use Leo. The Alice project, the late Randy Pausch's project, invited users to send electronic postcards. Something like this would be very useful for Leo. > Long before I talked about Scrivener[3] as a proposed > interface and the work of Terry stretching the interface possibilities > of Leo and Ville's and Kent's work on using and abstracting VCS, as the > talks about a one-click install and execution are putting the parts > together for this 5.0 release and the intend for a wider audience. For > me is quite interesting how non-programmers stick with Leo even with his > step learning curve. > > [3] http://www.literatureandlatte.com/scrivener.php Thanks for this reminder. Scrivener doesn't have clones, iirc, but it does some very good things. It's time for me to re-acquaint myself with it. > For me clones is what keep me on Leo and the potential of automatize > through scripts even if I don't use this. I agree. That's why I keep repeating that Leo's "internal" clones, that is, Leo's Dom and in-core data structures, will never change. Whenever you hear me talk about alternatives to clones, I am only talking about the format of external files, or ways of recreating internal clones without using gnx's. I'll be saying more about this today in another thread. > The abstraction of a VCS could > make Leo files travel with bazaar or git or fossil files so we could > have this outliner with his helper files in a companion VCS self > contained repository. > If I could thing for a single feature in the new > file format of Leo, this would be one that let people work > collaboratively on Leo. I agree completely, as amended below. > I have tried this before sharing Leo files in a > VCS, but it didn't work (reasons include that information about the view > of a file were stored in that file), but would be nice if a workflow > that let people use Leo to construct a shared understanding of a project > between Leo users sharing their trees in a collaborative fashion will be > the result of a format change. This is a very hard problem, apparently. My present opinion is that it is, in some sense, intractable. In the past I have considered many "clever" or even "heroic" approaches, but they all seem like bad engineering: complex and prone to fail in exceedingly bad ways. As I said above, my walk yesterday produced some truly new (for me) ideas. Rather than trying to do the impossible with .leo files, there may be a way to do more with @auto files. Again, I'll discuss this in another thread. > Cheers, > > Offray Many thanks for your comments. They are always thought provoking. Edward -- You received this message because you are subscribed to the Google Groups "leo-editor" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/leo-editor?hl=en.
