[TxMt] Re: TM2

[Allan Odgaard]

On 29 Apr 2010, at 22:28, Tim Harper wrote:

> […] Things have been awfully quiet about TM2 for the last several  
> months. You've claimed you are still working on it, but I've just  
> about lost hope. There are no signs to reinforce my faith that TM2  
> will ever become a reality.

This is understandable, and while I do feel bad about the situation  
overall, I dislike speaking about 2.0 — it’s a mix of many things,  
some I have previously mentioned, like not wanting to talk about a  
product which is partly just my wishful thinking of what I want the  
end product to be, another is that my motivation and optimism is  
anything but constant, sometimes I think that if I work really hard  
for two months, I have enough to release an alpha, yet there tend to  
be some stumbling block, something I estimated to just a day’s work  
(or didn’t consider at all) end up taking the two full months, so it  
seems I have made no progress at all after those two months (sometimes  
it may even go backwards because structural changes disable certain  
features), and someone will then ask about that alpha I was so  
optimistic about, which will make me think it is all hopeless…

So while the situation does bother me, I am content with people  
thinking 2.0 is vaporware and may never happen, because I don’t want  
to do posts of the day/week/month, as these will just be a reflection  
of my mood that day, which is anything but constant.

Pretty much everything I’ve said about 2.0, I have later regret. I  
wish I would have had the foresight to hire someone to have continued  
working on 1.x so I could quietly work on 2.0 — one may argue that  
this is not too late, but there is a better chance of me open sourcing  
2.0 than going back to 1.x. 2.0 currently has bad performance, lacks a  
lot of surface polish, has very few features people expect 2.0 to  
have, but it is a very solid code base which has good abstractions,  
most of the hard work has been done, and is geared heavily toward many  
years of future improvements/extensions/features.

So… 2.0 could go alpha but would go through same mocking and ridicule  
that 1.0 went through — with the letter I got from Aaron Swartz, it is  
something I am considering, but it will be pretty hard on me, and  
double-so because there are so many obvious shortcomings but I am  
still working from my own roadmap, so I wouldn’t really want any  
feedback, I mean, it’s like if you build a house, you don’t want  
strangers to come by and tell you that you need to put in a missing  
window while you are busy nailing the floor — so alpha release would  
solely be for users, but for most users 1.5 is probably still the  
better choice, but I will re-evaluate when the stories for the next  
three iterations have been completed (Pivotal Tracker speak), as these  
are the most glaring defects.

> […] I can't help but wonder if your promise of "free upgrades" has  
> blasted a crater in your motivation to work on TM2. Is there a  
> possibility of this, Allan?

Free upgrade surely was a mistake ;) But I tend to find motivation in  
other things than money, so it hasn’t really affected the timeline.

> […]
> * State that TM3 will be released with those features, and will be a  
> paid
> upgrade for everyone.

It will probably have to be the 2.1 release which will have to make up  
for the money I left on the table with that “free 2.0” promise ;)