[twdev] Re: Core testing and decision making

2009-05-07 Thread Morris Gray
On May 6, 7:38 pm, FND wrote: Thanks for taking my, sometimes editorializing, questions seriously. > These are good questions, Morris. > However, they're very jQuery-centric. I was thinking of a (still > somewhat vague) bigger picture - so in that context, jQuery might be > considered an implem

[twdev] Re: Core testing and decision making

2009-05-06 Thread FND
These are good questions, Morris. However, they're very jQuery-centric. I was thinking of a (still somewhat vague) bigger picture - so in that context, jQuery might be considered an implementation detail. (Nevertheless, the jQuery integration was a catalyst for reviving this longstanding issue.

[twdev] Re: Core testing and decision making

2009-05-04 Thread Morris Gray
On May 2, 1:43 am, FND wrote: > That depends. > If we were to embark on the 3.0 path without being encumbered by the > issue of backwards compatibility, I believe that could very well be > worth the effort of providing bug fixes for the legacy (2.x) line. That is a good insight and practical.

[twdev] Re: Core testing and decision making

2009-05-01 Thread FND
> I am reluctant to run two branches - doing so is a > significant amount of extra work and I'm not convinced that the > benefits it brings are worth it. That depends. If we were to embark on the 3.0 path without being encumbered by the issue of backwards compatibility, I believe that could very

[twdev] Re: Core testing and decision making

2009-04-30 Thread Martin Budden
Running two branches makes sense if we plan to do potentially disruptive changes on one of the branches. I read into your posting an assumption that using jQuery will, at least for a while, make TW less stable. jQuery makes TW bigger, but as long as we are careful, it should not introduce any extr

[twdev] Re: Core testing and decision making

2009-04-29 Thread Ton van Rooijen
Martin, In my opinion we would have, in SVN terms, a temporary period of 2 branches of TW. The first branch, let's call it "Stable TW", starts off from 2.4.3 and is only updated in case of serious bugs. No new functionality whatsoever. So over time there will be a new "Stable TW" called 2.4.4 and

[twdev] Re: Core testing and decision making

2009-04-29 Thread Paul Downey
On 29 Apr 2009, at 12:38, cd...@peermore.com wrote: > On Apr 29, 12:11 pm, Martin Budden wrote: >> OK, I understand this. What's your view when 2.5.1 becomes >> available - >> this does have added value (bug fixes etc). Should 2.5.1 then become >> the default upgrade? If not, what would be the

[twdev] Re: Core testing and decision making

2009-04-29 Thread Morris Gray
On Apr 24, 2:24 am, Eric Shulman wrote: > However, regardless of the elegance and technical correctness of any > engineering principles that are applied, these engineering > considerations should be, in my view, secondary to the needs and > interests of the TiddlyWiki end-user/author community.

[twdev] Re: Core testing and decision making

2009-04-29 Thread cd...@peermore.com
On Apr 29, 12:11 pm, Martin Budden wrote: > OK, I understand this. What's your view when 2.5.1 becomes available - > this does have added value (bug fixes etc). Should 2.5.1 then become > the default upgrade? If not, what would be the criteria for changing > the default upgrade from 2.4.3 to 2.5.

[twdev] Re: Core testing and decision making

2009-04-29 Thread Martin Budden
OK, I understand this. What's your view when 2.5.1 becomes available - this does have added value (bug fixes etc). Should 2.5.1 then become the default upgrade? If not, what would be the criteria for changing the default upgrade from 2.4.3 to 2.5.1? Martin 2009/4/28 Ton van Rooijen : > > Martin,

[twdev] Re: Core testing and decision making

2009-04-28 Thread Ton van Rooijen
Martin, Thought I explained it already more or less implicitly. I think that for the ordinary user a new release should have added value, even if it's invisible (e.g. bugfixes are undoubtedly added value). TW v2.5.0 however has no added value whatsoever for the thousands, only added weight. Plse d

[twdev] Re: Core testing and decision making

2009-04-28 Thread Martin Budden
Ton, I'm interested in why you think 2.4.3 rather than 2.5.0 should be the latest stable release. 2.5.0 is identical to 2.4.3 except for the addition of jQuery, and so should be just as stable - indeed it was designed to be a stable release users could downgrade to if they experienced any problem

[twdev] Re: Core testing and decision making

2009-04-27 Thread Ton van Rooijen
Although I am not completely sure if this is the right thread, here are my 2 cents: The "ordinary" user of TW, the ones that just use it as their scrapbook or GTD-tool, as a blog or as a website, or whatever TWs are used for, will in general not be interested in the latest and greatest TW version

[twdev] Re: Core testing and decision making

2009-04-24 Thread FND
> I wonder if it might be worth considering TiddlyWiki 2.x as stable, > give it a maintainer [...] with regular bug fixing releases. > And send all these jquery hopefuls off on a 3.x path that is > considerably more radical and accelerated than the current plans. Similar approaches have been disc

[twdev] Re: Core testing and decision making

2009-04-23 Thread cd...@peermore.com
On Apr 23, 5:24 pm, Eric Shulman wrote: > I think that we are really dancing around some practical, as well as > philosophical, differences of opinion related to the overarching > question: > >    "Why, when, and how should a given core change be made?" I agree and I'll take it a step further: