LeafLabs Garden

A place to share, learn, and grow...

1. 

   LarryP
   Member

   I read in the leaflab blog that the IDE for programming Maples was heading toward Wiring. On first glance, I don't offhand see much benefit. That may well be due to the IMHO poor Wiring documentation: Many, many fragments, and not an overview or architectural summary anywhere I could find. (IMHO, even by itself, unclear docs tends to indicate a less-than-clean design lurking.)

   Would somebody wise in the strengths of Wiring (and/or on the leafLabs team) explain why this (forkish) change in direction is a *good* thing, rather than a confusing, compatibility-breaking thing? IMHO, if an alternative to the Arduino GUI is to chosen, I'd much rather see leafLabs moving to a well-supported Eclipse environment.

   Comments?

   Posted 4 years ago #
2. 

   robodude666
   Member

   If you're currently using libmaple, or the MapleIDE, it does not affect you.

   Wiring 1.0 is a release that attempts to decouple platform-specific support from the language. In reality, for most people all it changes is... the IDE will be orange instead of green. That's about it. Leaflabs is working with Wiring to have libmaple built right in, and it aught to work the same or better than the current MapleIDE (from what I understand).

   If you're currently using libmaple from the command line, you can continue to do so.

   Posted 4 years ago #
3. 

   poslathian
   LeafLabs

   right you are.

   Also, an eclipse plugin would be nice. I have seen several Maple-eclipse setups that were built by hand using the command line tools environment we provide. It would be great to really polish that up and put it out there.

   In terms of Wiring, the exact path forward it still being worked out so I cant say too much. The long term vision is to provide a single environment with cross-vendor support and to develop a wirish (digitalWrite, Serial1.print, etc) level (rather than libmaple level) API that is standard across many devices, including crossing the 8/16/32 bit boundaries. This is a difficult undertaking to get right, so its really just getting started. The final perk would be that rather than a bunch of vendors all forking their own IDE, they can all put their efforts towards improving one IDE which will ostensibly mean a lot more polish. For example, the windows installation process on win7 64 -bit could become a double-click sort of endeavor ;)

   There are not a ton of finalized details for me to make available at this point, but we more than welcome your feedback.

   There has also been talk of an open shield standard, which I think is really important. Although IMO gadgeteer's peripheral model is so much more useful than the Arduino style shield. I dont know what the shield standard will look like, but it would be so great to be able to design a shield that works on many different vendors boards (and board types) and indicate that fact to all your customers. Not only that, but a standardized Wiring API (The Wiring Framework) would mean that any libraries/drivers written for a shield that adhere to the API would work across a ton of different devices. Its certainly a dream, but I think a really powerful one.

   Posted 4 years ago #
4. 

   tochinet
   Member

   For a good reason to migrate to Wiring, look at Arduino : they started from ... Wiring, made minor changes, up to 0022, and then ... broke it into the 1.0. Or look at Pinguino : they reengineered Wiring in Python for some unknown reason, hoping some day to support ARM cortex , AVR and PIC altogether (both 8 and 32 bits), and ... never got there.

   Maintaining an IDE is work, and it's a shame (IMHO) to have 4 or more parallel versions of nearly the same thing, just because of teh NIH syndrome. Some other people (Jeelabs, chipkit or pjrc's teensy) could understand that and wanted to "stay compatible" with the Arduino IDE. 1.0 was a pain in the *** for them.

   So I applaud the move of leaflabs to join forces with Wiring team. The more the better. I only wish leaflabs or wiring active people could give more info on what's going on (or why it's stalled), what are the plans, etc. Because I've seen little more than a LOI at this moment.

   BTW @LarryP, could you tell us where the Wiring IDE documentation is _most_ behind Arduino's or Maple's ? It looks similar to me, except there are less libraries.

   Posted 3 years ago #
5. 

   Rei Vilo
   Member

   @tochinet

   I fully agree with you: the NIH syndrome is devastating.

   For that reason and because I'm currently using different platforms —Arduino, chipKIT Uno32, ultra-low-power LaunchPad, Wiring and soon Maple—, I'm tired of having one IDE per platform, which only difference is the colour —blue, dark red, light red, orange and green.

   So I've developed a set of makefiles I could use with any IDE, in my case, Xcode because I own a MacBook. I also tested it on NetBeans.

   This makefile-based approach is based on the assumption that the claim " Writing a sketch is not programming " is just fairy tale. I'm confident that considering a sketch as plain C++ code could ease development, with no hidden "dirty work" done below.

   I'm working closely with Hernando Barragán to promote this idea.

   If you're using a Mac computer or just for fun, pay a visit to http://embedxcode.weebly.com/ to learn more.

   Enjoy!

   Posted 3 years ago #
6. 

   mbolivar
   LeafLabs

   rei -- looking forward to less forking and less NIH.

   Posted 3 years ago #

RSS feed for this topic

Reply

You must log in to post.