AltME groups: search

I suggest cheating. Start with this cool round menu and say "wel, 
platform lacks it anyway :)

Why GOB isn't a great name


Gob is commonly-used word in Britain and Ireland  Your Gob is your 
mouth. Gob Shite is a particularly Irish experssion to be used when 
somebody is talking rubbish. To Gob on somebody means to spit on 
them. A Gob refers to a portion of flem ejected from the mouth through 
the action of Gobbing (Spitting).

Instead of arguing, I propose we spend our time doing something productive, 
like reversing caret-to-offset and fixing that long-standing issue.

of course we can't push Carl to rename it, but the truth it, that 
it sounds a bit weird :-) And we know that Carl cares for naming 
conventions, right? My proposition was - that if 'feel stays, we 
can stick to what we have - feel, face, facet ..... and if face is 
gone? Currently it is not a datatype, but an object, so my proposition 
was easy - juste let's doc that from R2 to R3, face became a datatype 
:-)

rebol already has some funny names, like 'attempt, 'feel, 'what, 
but if gob! has a negative connotation, perhaps it would be a good 
idea to look for something else. still, GOB is in family with BLOB 
and BOB and other data related words. What about GROB?

I agree with Anton on changing names. In fact, I have totally changed 
my mind about renaming REBOL or any parts of it. The reason is that 
it would make obsolete all the documentation, etc. It would be like 
throwing all that work away, and would cause terrible confusion. 
I was perhaps the most strongly in favor of a name change, but after 
thinking about it more deeply I've decided I was wrong. Also, I think 
that REBOL has been around long enough that a fairly long number 
of programmers have at least heard of it. And now that R3 is coming 
out, it is very likely that a lot of people will start to give it 
more attention. So this is not the time for change.

Louis - correct, and that is why I think it should be called face!, 
so the name would stay, it would just change its type from object 
to datatype :-)

but this is one word in R3... there are no docs yet on it. I would 
assume that the naming and design phase of R3 is not yet entirely 
complete. so if there is a time to rename things, this is the best 
time to do it.

Anton, better now when all that needs fixed is RT's internal docs 
and source than hundreds of uses in the wild.  It really is an unfortunate 
name.  And there are some obvious replacements (over which there 
is no point in squabbling, we will not sway any decision).  Personally 
I think cell! or cel! would fit the bill.

a gob is just a graphic object, it does not contain a feel for example, 
and so on.

so a button could consist of 6 gobs, one for each edge, one for the 
background and one for the text?

because the C struct only has one "slot" that can either contain 
an image, or a draw block, or a text block and so on.

yes, it is what is PUSH for, no? But that concept imo collides here, 
because we have now separated draw, effect, image. But maybe it is 
not. I had some more difficult UI set-up in mind, and the ability 
to save certain rendition without the need to constantly regenerate 
it. But maybe that is done by compositing engine itself ....

hmm, now thinking about simple thing as red text - will it use two 
gobs? text and color are mutually exclusive ...

if you mean background - then i think so, you have a bg gob and a 
text gob inside or over it.

sounds scary ... as ZX spectrum, which could have only 1 color for 
particular bog pen, and one color as a background, where that gob 
was 8x8 pixels :-)

this is done for speed and saving memory

ie it's faster and uses much less memory this way.

what I don't like (but the design is still not finished so this may 
change) is that there is separate draw and effect block. I think 
R2 way (draw in effect) is much more flexible. Probably merging those 
two dialect would be best.

rebolek, most of the times you are not using both. and using layered 
gobs is going to be faster, probably.

they would all have the same offset and size then :)

Carl: "Ask if they remember what great graphics system used "boogers 
and goobers".  ;)"

Carl: "If we had to pick worldwide perfect names, we'd have very 
few to pick from.


Also, when you define a word, it is context dependent. It very rapidly 
takes on its new meaning and loses the out-of-context meaning.


Also, if they can create a better 3-letter name that starts with 
G -- then I am open to hearing it.  But, it must happen very soon."

i'll wait a day or so then collect proposals and send them to Carl. 
don't keep your hopes to high though ;)

VIS - view image structure , and the french word for face :)

well, if face is about to stay as highere level organisation of gobs, 
and we have facets, feels, what about eye? :-)

well, we have amiga devices, many amigans would be familiar with 
sprite, and general publicity with layer, others are pretty much 
cryptic ....

but gob is not an object :-) Well, but I get your point. IMO Carl 
let us talk for a few moments and then we will realise that we stick 
to gob :-)

both. Gremlins are the cuter ones (if defined by movies).

About gob, i imagine one speaks with customers about the gui and 
uses gob all the time. Sounds not that nice..

and evil of course.

well, and many of GRO is Gros? Our former prime minister :-))

Sounds good. and http://en.wikipedia.org/wiki/Groo_the_Wanderercoud 
count as an insider, its graphics related too :D

I don't like gob; gfx and vis seem like better choices to me. Carl's 
limitation of three letters starting with a G limits us a bit. Under 
that rule, I would probably choose gfx.

So far proposed words:

(gob! face! gfx! cell! atom! layer! vis! sprite! bob! blit! )

the only which really applies is gob! or vis!  others are all pretty 
much non-related.

people you forgot the most obvious and pretty!

gel!

graphic element.


but I don't see the point of the heated debate over the name gob. 
 gob is chunk, a morcel of something.  in this case its pretty precise. 
 its a part of the gui, not the whole gui... and in another sentence 
its also meaningfull... when you say gobs of text, it means a lot 
of text... so in reality we WILL need gobs of gob to make up one 
gui.

(sorry I forgot gfx which is also related (and my favorite appart 
from gel! or gob!))

Where it says: Grab the gel and drag it around.

I would like to know, how and if /library interface changed for R3. 
It was long promissed, that integrating external resources should 
be made easier (see Carl's answer in RT QA channel IIRC), and I could 
not find any mention of it in Carl's DevCon slides. I would excpet 
changes in at least how arrays of chars, structs are handled, as 
well as more than 16 callback support .... maybe even more datatypes 
supported. 


Maybe a rather weak state of library interface put REBOL into integration 
isolation. It is not easy enough for newbies to wrap to libraries 
....

one thing that I would love for R3 is the ability to directly access 
the image byte data of a rendered window - a read only ability would 
be fine - or even editing with a lock/unlock mechanism.  The main 
reason for this is the allow direct pixel analysis for non-standard 
interfaces - eg AGG/Draw based interfaces where colour can be used 
to determine function of an irregular shaped button.  The basic premis 
is that you draw your UI composite then rather than having to do 
cumbersome bounds checking you can simply test the pixel colour under 
the mouse and call an activation function based on that colour - 
eg 0.0.254 = a save function.

Perhaps a simple way to do this would be to implement double buffering 
by default and allow buffer access.

at least with a core and async I can get the MMORPG server engine 
back up and running...

how /library changed - no change yet, i see library! and routine! 
in R3 but i assume that is still using the R2 code... (or not, but 
Carl has not documented it yet). I think, this will be part of the 
open source code, so you should not need to worry.

So, who's been a bad boy then?  We can eliminate those and figure 
out who the good boys are :)

Graham... hmm... you used logic in some discussion, that is a very 
bad thing to do, as Terry and Louis teach us.

hey, I've been adopting santa as a religion over and over  :-)  I 
even tried to convert terry last week!

Gas is too formless, and everyone has some gas that immediately comes 
to mind. It kind of stinks as a name.

I guess I'm not a select developer :(  No more using logic to talk 
with Louis and Terry !

I have heard that you, Cyphre and Carl had some chat about new VID? 
Is there any consensus? Or, path-to-follow defined? Will it be your 
proposal implemented?

we have ideas, we'll document them and we'll make sure we'll get 
to a vid that people will like.

nope. but we'll work closely with ashley. we hope r3 vid will be 
better than rebgui, but diversity is a good thing, and i'm sure some 
people will prefer rebgui.

Carl did VID as an example and we wanted it being perfect. And only 
lately it was Ashley, who understood situation, and produced separate 
GUI

I can imagine R3 could bring diversity to the GUI. I can imagine 
e.g. GUI for Presentations (as Carl's script) and also GUI for kind 
of Scala product ... we can build Scala "killer" based upon plug-in, 
with various transitions, etc. For such things, something like RebGUI 
is not suited at all.

Graham - I think it is in planning, we call it generally here VID+, 
and that it is going to be a rewrite ...

we will improve it afterward of course, add more styles, and so on.

Graham, I think that it is possible. The engine and subystems are 
not so big imo. And we are in different situation a bit - we have 
ppl knowing VID and its weak places really deeply ...

perhaps a hide cursor option and display a custom cursor for full 
screen apps like tv apps

hmm, and no fear Ruby guys will take it and link somehow to ruby? 
:-) Well, they would still need even rebol, but that is not bad plan 
either .... they will link View and Rebol, later on realising, they 
don't need Ruby at all :-)

they can already take AGG and link to it.

and it's good if they link to rebol.dll :)

Gabriele - they linked already to it! They have bounties for projects 
and they don't hesitate to call it Rebol/View clone! It took 3rd 
place. Then I saw, how author was disappointed, linking to AGG, how 
slow it was :-)

the interesting part is the rebol part - ie vid, the rich text dialect, 
the draw dialect and so on.

was thinking mainly for when we get an IDE and want to do code block 
folding etc.

Gabriele - re docs translation - that needs some organisation too. 
Imagine View doc. It is a very good doc, right? Now someone want 
to translate it, and he/she does so. Then someone does some tweaks/additions, 
changes to it. So - how translators are notified about it, to not 
actually need to look thru all doc for what changed?

rebhost.exe? Weird name. Hope it will become rebol.exe, once stable 
and fully released :-)

ah, name is just because this is the host for the dll :) ie rebol 
interpreted is an embeddable library... and it's always embedded 
into some host application

the public release will just be rebol.exe, without external dll and 
so on. developers instead can download the dll, and embed it in their 
app, or customize the default host for their needs and so on.

I just wonder ... dll is just cross platform library, right? And 
if I understand it correctly, then in that dll there are mainly structs 
defined, which you have to fill-in, no? IIRC something like that 
was said in blog regarding getting OS time. So I wonder - is linking 
dll to your app sufficient to run rebol code? Or you would have to 
port also platform specific things?

also, the host defines some natives... eg. browse to make an example. 
but of course show, and same will be for call etc.

I wonder if gaberiele ever feels like the fortune teller at a fair 
- all these rebolers coming up and asking about their future :)

R3 will eventually replace R2, but for a while both will be available 
and supported.

plugins - the alpha does not load them dinamically, but it's basically 
what the host program does (defining devices, natives, and so on)

and it will load them dynamically later?

R3 does not have encryption modules yet, they will probably come 
later. but notice that nobody stops people from linking to OpenSSL 
and provide a complete SSL solution.

not sure if correct, but I can see rebin as kind of "packager" for 
rebol? Will it allow us to pack various things - e.g. libraries, 
scripts, images, or other resources, and add it to the rebol.exe?

rebin is a binary format for rebol values. like now we have a text 
format (produced by mold), then we'll also have a binary format useful 
both for storing (when user-readable is not needed) and for internal 
communication with plugins

and, R3 might just use the same module found in R2 for SSL

SSL is specific protocol. I am not speaking about encryption here, 
but about the trust, that is what certificates are for. IIRC the 
api for them is inside, it was not just exposed. And IIRC2, Ladislav 
once mentioned he will do it? But I could misunderstood him ...

then you misunderstand, what is important, and what is not.

today's sw world, distributed, without signatures? The thing is, 
someone from RT confirmed that the stuff is inside, just not exposed. 
And that is waste of resources. So - expose them.

then you misunderstand, what is important, and what is not.

 -- I think Gabriele misunderstands very little. :-) The thing is, 
 it's all about compromises. What is important to you might not be 
 important to me at all, and vice versa. So how do you choose what 
 to include, and how big is too big for REBOL to be? 


If something is wanted and needed by 80% of the community, or 0.1% 
that's doing something really important, those are easier calls to 
make. Many features have doubtful value to at least some people, 
but we can't use those as justification for adding other things of 
doubtful value. So, I want a good voting system, and tracking for 
new community mezz funcs, to see what gets used the most; what people 
want and need.


That said, I think security is so important now that anything we 
can do to make REBOL a better tool for writing secure systems is 
a good thing.

But I am not talking about new features at all. Someone said, that 
Core contains certificates handling already, which means, even parsing? 
It is the same situation, how Rebol contained internally convolve 
function, which was just not exposed. So, we are talking nearly zero 
addition to Rebol. And not having certificates handling in rebol 
directly is what actually stopped rebol plug-in. If I'll see some 
reblet signed from Gregg, I will not think any second to just press 
"Run". 


I am not now opting for the particular functionality - my question 
was more general. Plug-in development stopped, because Josh stated, 
that he is working on soon-to-be-released new schema for Rebol security. 
We all knew how it would end, and we were not wrong. That is exactly 
the reason, why I ask - there is no plug-in, without the security. 
And plug-in is imo very crucial product now.

Forget certificates right now, before we enter another pointless 
discussion. The thing is, that I believe RT regards reblet signing 
important thing, and that it will come. If it is not part of the 
system, noone will do it, period. The same reason, why RT pushes 
for rebservices - it will be standard without any need for further 
discussion, if we do it this way, or that way ...

you're still confusing signing with certificates. the ssl module 
certainly has to parse certs in the ssl handshake. but that does 
not mean that it will parse any cert file format. and, there are 
many. but... even if it was exposed, what would it save you? three 
lines of parse?

I am not sure I am confusing anything. If it is not there, then it 
is not there. I can give an example - Bobik - he left rebol, because 
of our often claims, that we can do anything. But he is looking for 
the end user tool. He does not want to code mySQL driver himself, 
he wants to use one. So, if you will have website with rebol features, 
what will be your answer to following bullet:

certificates support: 


Yes, or no? Am I able to easily send rebol email, signed, which displays 
in Thunderbird or Outlook, as signed? Am I able to choose from centrally 
installed certificates in Windows certificate container? That is 
my point. So - if it is only 3 lines of code, just take a note, and 
when RT will be thinking of security/privacy issues, please count 
such things in?

Because last year, I was asking about certificates. There was a possibility 
for me, to have small rebol app, which securely sends and checks 
documents. It could be used for invoices exchange. I stopped because 
noone was able to point me out, how should I check for signatures.

Whereas guys from Delphi crowd were able to do that. Now you can 
blame me, that I was not able to make it. REBOL nor its community 
did not work as enabler for me here. And I can see only one sensible 
way to avoid that in the future - projects domain, bounty system 
...

and worst case - why not use CALL to call openssl??

Gabriele - you constantly provide the same picture, and if we guys 
don't change attitude, we will not get new ppl attracted.

You could do that in one hour, for me - I did not find any help, 
and it was show stopper for me.

we don't want to turn into perl, that's my only concern. otherwise 
we'd just go to perl and have all the new people you want.

now... why not having x509 parsing in rebol? that's surely possible 
and probably a good thing to do. but, tomorrow you'll find something 
else that's missing.

I remember exactly the same discussions when Terry was proposing 
RASH - we claimed we could do everything Flash can. Yes, in teory. 
But there is many ppl, who look for new tools in internet era. Some 
of them, don't want to code drivers etc. themself. In fact - they 
choose upon what is available. To avoid this situation, I propose 
bounty system, it would be vital. So guy like me could take some 
money and sponsor some development.

and why should RT write all the libraries?

how much would it take to extract the key and signature from that?

having more time, i'd just study the binary file format and parse 
it myself. i don't see any show-stopper.

but again, assume this was built in, or available as library from 
rt. you would simply find something else that is not built in and 
complain about that :)

OK, I will ask the other way? How would you trust I am me? (or how 
to write it in english), if not to trust some third party = CA? IIRC 
Carl or Josh said, that new SDK (dunno if it was ment generally, 
or only for plug-in), could generate some special license key or 
certificate for me, so that users could check? Similar aproach as 
when you install driver and you have the ability to check, if it 
comes from trusted party. That was all my point why I asked if it 
is coming.

since i need to be sure that the public key i have i yours, and not 
someone else, we need a trusted party - that's called a CA

ie someone that certifies that you are actually Petr and not an impostor

I know - but do you think ppl will trust some home made CA? I thought 
that I will come to one of three CAs here, and will ask for commercial 
certificate (dunno who does so internationally - Thawte? Verison?)

and if you don't even want to trust rebol's internal rsa code... 
you can call openssl for everything. that code is trusted by all