mailing-list for TeXmacs Users

[Magnus Rohde <address@hidden>]

Den 29/9-2005, kl. 15:10, skrev Henri wrote:


>  criticism is in fact very harsh ; this is not forbidden : but then, 
> in order to be meaningful, you should be able to point to precise 
> points ; just saying "doesn't come close to any drawing program" 
> without giving details of what you really mean is useless : I cannot 
> use such a very vague information to improve the software.
>
> So as far as Xfig is concerned, I would say that, to use your own 
> words, we come quite close to it : the current version of the graphics 
> mode of TeXmacs contains most of the features that can be found in 
> Xfig, and it is implemented in a much more user-friendly way.

As I said I only mentioned Xfig as an example of a drawing program. I 
chose it (perhaps mistakenly) because it is widely used, I have never 
used it, and I probably never will.

> .....
>
> Given these facts, I think I clearly showed why, in my opinion, it is 
> **today** that we can say that the graphics mode in TeXmacs comes 
> close to Xfig (I discuss specifically Xfig here : of course, I do not 
> claim that we come close to any drawing program). So my question is : 
> why do you think that the current version of the graphics in TeXmacs 
> would fail miserably, if one compares it to Xfig ? (be specific, 
> please).

As I have never used Xfig, I cannot compare the two, but when you 
compare TeXmacs to any other drawing program (eg. OpenOffice Draw) you 
notice several things:

(1) You have to use more than one mouse button to create a line. It may 
seem inefficient to only use one mouse button, but that is the way most 
programs work and it is therefore the most familiar. (actually this 
goes for the entire interface)

(2) No colour gradients or transparency.

(3) I cannot import and place graphics from other sources (pdf files 
etc.)

(4) There is no way of knowing what function the mouse is performing 
(will I move/create/assigning properties to this object) or what tool I 
have chosen.

(5) There is no align/distribute function.

(6) There are no custom fills.

The most important here is (1). If I need more that one mouse button 
it's too complicated.


> Money is not relevant for defining what value is. Rather, value is 
> defined in the first place, and money is then used as a tool that 
> allows people to invest in this or that particular value.
>
> In other words : the fact that some software is free and some other 
> software is not free is irrelevant : it doesn't free us from the task 
> of explicitly saying what we want. Saying "this software, being driven 
> this way or that way, offers real value" is not even an argument : it 
> is politics.

I wasn't the one who started discussing politics, that was Alvaro when 
he said:

> On the Illustrator point I only have to add that it is rather the 
> opposite way
> to the one you put it. It is impossible for Illustrator to compete 
> with any free
> application because it is not free.

I consider software as tools, and if a tool cannot get the job done it 
offers zero value for money, even if the tool is free (like beer or 
speech, doesn't matter).

> On the other hand, by stating that the feature A is missing, or that 
> the feature B should, in your opinion, be modified this way (the said 
> way being described in sufficient detail), you do something useful, 
> because you bring explicit, clear and logical arguments that people 
> can understand, reuse, and improve. If nobody ever does this, no 
> wonder that the development of TeXmacs is driven by the developers, 
> because without such a task of turning problems into explicit 
> (possible) solutions, no software development is possible !
>
> > I know that Joris does put many hours into the project, and my 
> > comment wasn't on his or any other's work. What I was trying to say 
> > was that I don't think that the direction of development should be 
> > guided by the interests of the available workforce.
> It is not a matter of available workforce : it is a matter of writing 
> specifications. If you want that your ideas become implemented, you 
> must first describe them in sufficient detail, otherwise it is 
> **impossible** to take any ideas into account, because unfortunatly, 
> the current "available workforce" has no telepathic abilities yet.

My original comment wasn't  on specific features, but rather on the 
general direction of development (the big picture or the whole 
scientific office thing). However, this discussion have given me the 
idea that a user survey would be useful for determining where the 
developers should concentrate their efforts.


Magnus