Re: [Scid-users] A Simple Step to Help TWIC (or any other Ad Site)

[Mark M. <cor...@co...>]

Nice talking to you.

On Sat, 2012-01-14 at 17:10 +0100, Alexander Wagner wrote:
> On 01/13/12 21:30, Mark Morss wrote:
> 
> Hi!
> 
> > Since you ask, I perhaps should say that I am an old Unix hand and I
> > fully subscribe to the "do one thing well" philosophy.
> 
> Me too.
> 
> > Much of what I
> > see from the SCID list server causes me to worry that SCID is on the
> > path to becoming bloatware.
> 
> I "fear" you're pretty safe from that. Actually Scid should get a lot of 
> polishing but not new features at the moment. Unfortunately however, 
> I've currently nearly no time to work on Scid myself. I'd have a bunch 
> of things on my list, however.
> 
> > With specific regard to automated downloads from TWIC or anywhere else,
> > I would much rather see this delivered in a separate executable than as
> > an integral part of SCID.
> 
> Perfectly agree. And to assure you: we will not add such a function. I 
> asked Mark if he'd like to have something like that, he doesn't, we'll 
> not do it.
> 
> > I hope those responsible for SCID's direction take into account that
> > more and more "functionality" does not always add to the utility
> 
> He does :)
> 
> > of a piece of software.  From this point of view, ChessBase and Chess
> > Assistant are very bad models to follow.
> 
> Agree.
> 
> > In line with this, I find the thought of links in the game headers,
> > which would undoubtedly call some sort of attention to themselves,
> > pretty horrifying.
> 
> Probably, this would depend on how it is handled. If you just place 
> something like
> 
> Source "TWIC
> 
> there where you could click on the TWIC you'd hardly notice it. But to 
> make sure that you've not to be concerned about it to happen:
> 
> 1) Someone would have to do it in real life code. Experience tells me 
> that chances are not good for this to happen.
> 
> 2) The devil lives in the details here: What would you link to? It would 
> only make sense to link to some sort of URI (persistent URL) as 
> databases, aka _archives_, tend to live quite some time so all URLs will 
> lead to nowhere in almost no time (compared to the time scale of a 
> database). So, as long as chess games do not get URIs assinged that make 
> sure that the source it links to is available next year, it doesn't make 
> sense to store any links anyway. You may get back in the list archive to 
> a lengthy discussion about the feasibility of such links, how one could 
> build up URIs on basis of persistent identifiers and so on...
> 
> > If somebody wants this, it should be his burden to switch it on, not my
> > burden to switch it off.
> 
> The only way to handle it decently would be like the outbound linking we 
> currently do in the players information. Note that this is indeed based 
> upon persistent IDs for the players, and as such only available for a 
> comparatively small number of them. (Check out Anand or Kasparov e.g.)
>