Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
  • Mar 23 2017 22:39

    greggirwin on master

    Update map syntax (compare)

  • Apr 18 2016 21:46

    greggirwin on master

    Forget about html in MD as well… (compare)

  • Apr 18 2016 21:44

    greggirwin on master

    Forget about MD lists in tables. (compare)

  • Apr 18 2016 21:43

    greggirwin on master

    Table test 3 (compare)

  • Apr 18 2016 21:40

    greggirwin on master

    - Fixed example comment alignme… (compare)

  • Apr 18 2016 21:37

    greggirwin on master

    - Added goal/use table test. - … (compare)

  • Apr 18 2016 21:30

    greggirwin on master

    Added top-line goals and uses, … (compare)

  • Jun 10 2015 18:57

    greggirwin on master

    Added Gitter badge Merge pull request #1 from gitt… (compare)

  • Jun 10 2015 18:57
    greggirwin closed #1
  • Jun 10 2015 17:07
    gitter-badger opened #1
Nenad Rakocevic
@dockimbel
:point_up: June 11 2015 2:47 PM With simple quotes now, is it ASCII?
Christopher Ross-Gill
@rgchris
The degree symbol (°) is beyond Ascii (U+00B0).
Nenad Rakocevic
@dockimbel
:point_up: June 11 2015 2:49 PM Would that matter in practice?
Christopher Ross-Gill
@rgchris
It's not an easy symbol to reach either.
(Opt-0 on US-English Mac—perhaps that is easy enough, don't know about other keyboards)
It does seem like a significant decision though—to accept delimiters beyond Ascii.
Would we accept strings delimited with curly quotes? (U+201C, U+201D)
WiseGenius
@WiseGenius
I would! But that's just me.
Is immediately preceding digits with a - a wise choice of separator?
Nenad Rakocevic
@dockimbel
:point_up: June 11 2015 2:53 PM I just realized that ° is not part of standard QWERTY keyboards (while it is standard on AZERTY ones French people use). :-)
:point_up: June 11 2015 2:58 PM Obviously not. ;-) Well, degrees values in such case cannot be negative, so it would be fine. ;-)
Christopher Ross-Gill
@rgchris
June 11 2015 3:01 AM A few degrees of separation there :D
WiseGenius
@WiseGenius
:point_up: June 11 2015 5:01 PM Yes, they cannot be negative, but I still feel it's a poor form. Also, other notations often use + for N and E and - for S and W, and that could be a source of confusion.
Nenad Rakocevic
@dockimbel
:point_up: June 11 2015 3:12 PM Sure, a much better format could be designed. Given how geo-coordinates are ubiquitous nowadays, having a good first-class support for them looks like a useful feature to have. If anyone would like to design such new datatypes (syntax + semantics) for Red/Rebol, the github wiki awaits you. ;-) I would certainly be glad to add it to Red once we have a proper spec.
We can always sell AZERTY keyboards with a Red logo on it for good support for typing °. ;-)
WiseGenius
@WiseGenius
If it were customizable, I'd be first in line to buy one! :)
Nenad Rakocevic
@dockimbel
:point_up: June 11 2015 3:22 PM I have considered such business model for our company, though, we would need a user base of a few hundred thousands at least to make it profitable enough. So probably not for this year. ;-)
Gregg Irwin
@greggirwin
https://github.com/Ren-data/Ren/wiki/Gitter-Room-Index (as I figure out how to organize Gitter stuff @PeterWAWood )
Gregg Irwin
@greggirwin
I won't add people to rooms now, since I think they can find them. It's a pain to add lots of people individually. Can't copy them from another room.
iArnold
@iArnold
:point_up: June 12 2015 12:54 AM Perhaps the notifications could be as a standard be set to 'no'?
Rudolf Meijer
@meijeru
:point_up: June 11 2015 8:20 AMThis issue was reported under #1213 and has been corrected.
WiseGenius
@WiseGenius

:point_up: June 12 2015 4:50 PM By red/red#1213 being “corrected”, do you mean that the compiler now errors too?

*** Syntax Error: Invalid word! value
*** line: 3
*** at: "1..2^M^/"

The latest build's interpreter still gives:

*** Syntax error: invalid value at 1..2

I wasn't complaining about the result, but was just using it to clarify that tuple!s don't have “blank” parts, but 0s.

Rudolf Meijer
@meijeru
:point_up: June 12 2015 9:14 AM The correction was meant to ensure that empty tuple elements would be allowed both in the compiler and the interpreter. If you find a different behaviour, that is a regression and should be reported as a new issue.
Andreas Bolka
@earl
This message was deleted
:point_up: June 12 2015 2:15 PM The actually implemented correction ensured that empty tuple elements are allowed in neither compiler nor interpreter.
red/red@fa2ef36
Rudolf Meijer
@meijeru
:point_up: June 12 2015 5:38 PM I stand corrected ;-) Compile time and run time lexers are in sync, but this is now a clear departure from R2/3 on this point.
Nenad Rakocevic
@dockimbel
@greggirwin I don't remember in which group you mentioned the option of creating a "mission" group for Ren, I think that would be helpful. I am in favor of a minimalist approach, that can be expanded incrementally (as I think that is the best practical way to achieve it) rather than an all-encompassing approach from beginning, that would raise the bar too high for the implementations across languages (including Redbol).
Gregg Irwin
@greggirwin
Gregg Irwin
@greggirwin
I dumped some notes in the above Mission doc and added a Votes section to https://github.com/Ren-data/Ren/wiki/value-point-rfc as a possible way to track consensus. The idea of the Date column is a simple mechanism to know what version of spec a vote is attached to.
Gregg Irwin
@greggirwin
@asampal , look at the Wolfram Data Framework and see if that's what you're thinking for specifics and general tuple use. That's what I think could be built with Ren.
And build it wikipedia style. Too much work for just us. ;)
Boleslav Březovský
@rebolek
So Humanistic was abandoned and this is new home for Ren, until there will be decrease in activity and we create something new again few years later?
Adrian Sampaleanu
@asampal
yes, why didn't we go with Humanistic?
@greggirwin , you're an owner there too, no?
Nenad Rakocevic
@dockimbel
Humanistic name looks much cooler than Ren-data. ;-)
Gregg Irwin
@greggirwin
We can maintain both in parallel if you want. Humanistic is a fun name, Americanized, and I personally like the Humanist reference. Ren-data is boring, practical, and says exactly what it is. Which is better for helping Ren succeed?
As we add more projects, Ren-data is also more constrained about what should be under the organization's banner.
Adrian Sampaleanu
@asampal
@greggirwin , at this point when things are just getting going wrt Ren, I would think it's needless confusion to have more than one official location.
Gregg Irwin
@greggirwin
OK. My vote is for Ren-data. Since I'm not sure of how visibility works WRT github organizations, I created a new team and invited some people from here. Nobody was intentionally exlcuded, just based invites on activity here and past interest. Now at least I think people should be able to see what's there if they couldn't before.
I haven't moved my ren-data.org stuff there yet, but still plan to.
Gregg Irwin
@greggirwin

I put up https://github.com/Ren-data/Ren/wiki/value-none-rfc and a template for value RFCs. Is this kind of thing worthwhile, or does it add too much formality (and work)? For simple values, it doesn't really help. The doc could just say what I have on ren-data.org

The word none means nothing. Rather, it means something could be there, but nothing is. It isn't zero, that would be null, but Ren doesn't use null; it uses the number zero to mean zero. None is like nil or nothing. When the choice was made, nil nothing none of the other options sounded as good.

That's all the informative info we really need, and the grammar is the normative part. The only reason to do this for simple values is completeness, because we probably do want this type of doc for complex types.
Christopher Ross-Gill
@rgchris
Mark is working on the Rebol lexer related to Fork's endeavours, posted this brief proposal on tags (called Tags III): http://chat.stackoverflow.com/transcript/message/23678540#23678540 (not an endorsement).
Gregg Irwin
@greggirwin
Cool. I have an experimental R3 Ren parser I'm using to test ideas and find obvious grammar conflicts.
On Ren/Strict, the first variation on Ren I might suggest in MinRen (Minimal Ren), which would basically be a slightly nicer JSON. Still thinking.
I haven't read the whole transcript, and I haven't yet included tags in Ren. My original thought was that they are just another syntax for multiline strings. One edge case being empty tags.
FLuX LoOP
@x8x
Transit is a format and set of libraries for conveying values between applications written in different programming languages https://github.com/cognitect/transit-format
Gregg Irwin
@greggirwin
Doesn't grab me at a glance. What do you like about it?
xaduha
@xaduha
I think that a data language is essential to have, I don't even care that much about details, as long as it's there as a first-class citizen and can be easily marshaled to JSON.