Fog Creek Software
Discussion Board




Knowledge Base
Documentation
Terry's Tips
Darren's Tips

Get a move on, FogCreek!

So I caved in and bought a personal edition of CityDesk because my trial version locked me out.  But I'm warning you, Joel, I'm shopping for server space that supports PHP, CGI & SQL.  Once I find a good deal there, then that opens up my options.  The price is right for Movable Type or PMachine.

Release CityDesk 2.0, and have Undo work consistently across all Windows OSes (Your Undo in XP Home is time sensitive), improve the usability of swapping views between Normal and HTML view (don't overwrite our hand made code, even if you have to cache the whole page away), and show us a richer scripting environment (prev & next links).  Please keep my money.  I don't want to have to ask for a full refund.

David Blume
Friday, March 14, 2003

Wow, even the new customers are irate.

www.marktaw.com
Friday, March 14, 2003

Not irate.  Anxious.  It's time for version 2 to come out with features to work as we *expect.*

Remember when CityDesk saved and *closed* the document when you hit Ctrl-S?  Ugh.  I still think the "Save and Close" button should just be a "Save" button.

Even "Find" is not intuitive for me.  I'd prefer it to work more like MSDev than MSWord.  (Regarding Finding within a selection, wrapping and direction.)  I understand I may be in the minority on this one.  Hope not.

Anybody here find it interesting that HomeSite 5.0 removed the "Designer View" it had in 4.5, and it's better off for it?  Different audience than CityDesk's, I'm sure, but that's how frustrating it was to have hand-coded text reformatted when switching views.

CityDesk is potentially fantastic software, and I'm anxious to put it up there with Excel, FeedReader, ExamDiff Pro, CVS and MSDev.

But I'm frustrated that I paid real money for potentially great software.  I want the great stuff in hand.

www.daliblume.com
Friday, March 14, 2003

I never used Homesite's designer view. Too bad it was basically replaced by Dreamweaver.

www.marktaw.com
Friday, March 14, 2003

I just checked here out of curiosity.  Man, commercial dev is such an antagonistic process.  May I be blessed with never having any customers. ;)

Of course, entertainers complain of the same problem.  That's why entertainers generally make more money than artists, and why artists fear being entertaining...

Tj
Saturday, March 15, 2003

I'd rather wait for quality than end up with some half-finished rushed job.

John Topley
Saturday, March 15, 2003

What surprises me most about the "hurry up and ship Fogcreek 2" comments are that some of them are made by software developers. Surely they know more than most that the release cycle of a piece of software (development, testing, bug fixing, testing...) negates releases less than every 6 months or so on a product of CityDesk's size with the number of people I suspect are working on it. And of that 6 months you can only really spend half of that (if you're lucky) actually doing new development.

With a small team you've really got to try and pick how much time you spend on bug fixing and new development, balancing the time a bug fixing release cycle costs against the cycle of a new release with new functionality.

The bottom line is that no amount of nagging can magically change the rules of releasing software and it won't come any sooner than when it's ready! Patience is one of the hardest things to learn in life, but it's one of the most valuable (imho). Don't get me wrong though, I'd jump at the chance to get my hands on v2...

John C
Saturday, March 15, 2003

Hear, hear to the comments from both Johns.

David Burch
Saturday, March 15, 2003

I am a software developer too (or at least, I am into consulting, and do some research on systems engineering methods). So here's my 2 cents:

Time boxing is a very interesting way of controlling projects as well... I have learned that if the release cycle takes very long (and a year is very, very long i.m.h.o.!) that the project is less likely to reach stable baselines. It needs to be changed again and again and is never finished. (still hope this is not the case for CD2). I'm not in favor of lousy quality early released, but I'm willing to take less functionality for early release and see it increase every 3-6 months...

Adriaan van den Brand
Sunday, March 16, 2003

3 - 6 months might be fair for consulting projects or in-house development tools, but when you're developing a product to run in the wild that's peanuts. Even just doing a modicum of beta testing takes 3 months. By the time you've got yourself a task list and done the design it would leave about 3 hours on a saturday afternoon to cut the code.

I've been doing a lot of development lately on what are ultimately custom consulting projects but which need to run 24/7 in hard-to-get-to production environments. It's been a real eye-opener for me (and really satisfying too) to be writing much better code because of the extensive testing we do. But it also takes about twice as long as the rush-jobs I've been used to in the past.

I for one (and a few others by the sounds of it) would much rather wait for a good CD2. I know there are a few problems with v1, but it does me fine thanks.

Sam Strachan
Monday, March 17, 2003

I've been involved in a lot of successful software projects over the years (Delphi, JBuilder, WebMethods etc) and one of the things that is obvious is that markets don't wait.  Frankly, I think anything more than a 12 month release cycle is likely to do more harm than good.  When we kicked off Delphi 1.0 in early 1995 (Feb 14, actually) we were able to follow up a year later with a full 32 bit version, new controls, better reporting etc.  We kept up the pace on a 12 month release cycle and I think it was one of the reasons Delphi did well in the market place.  That doesn't mean you can't have deep innovation.  We had a small team, but some developers worked on the current release, and some were working on things that would be in the next release. 

I think 18 months is too long for CD 2.0.  I've been using 1.x for about a year and I hope the new release is out soon.  But I encourage Joel et al to make sure the releases are more timely.  That doesn't mean rush something out in 3 months, but if you want to serve your market, you should listen to your customers.  Oh yes, communicating with them is also a good idea.  Otherwise, like it or not, they tend to drift away.

--Zack

Zack Urlocker
Tuesday, March 25, 2003

When Delphi 1.0 hit, it did so behind a massive wave of publicity. So Delphi 2.0 pretty much had to follow up on a lot of promises, or all that marketing momentum would be lost.

CityDesk 1.0, however, has only really been marketed by word of mouth. Joel has said before that he wasn't going to do much (any?) marketing until CityDesk 2.0 was ready.

So yes, the long wait for CityDesk 2.0 might annoy some of the few thousand people who've heard of or used CityDesk 1.0. But if it makes CityDesk 2.0 a more polished product, there'll be thousands of new customers banging on the door when they see it being advertised. If CityDesk 2.0 isn't well-polished, Joel will be disappointing many thousands of potential customers, and will have a hard time winning any of them back.

From that point on, I agree that more frequent releases will help.

Darren Collins
Tuesday, March 25, 2003

Darren, I agree these were different circumstances and different times etc.  However, you could make the same argument that CD is losing momentum given the lack of information forthcoming from FogCreek.  It may be on a more modest scale, but it is still happening.  Also, I think with a good software team (and I would put FogCreek in this group) the tradeoff is not whether a product is polished or of high quality.  The point is not to tradeoff quality.  A new release should always be of high quality (e.g. few bugs) but I am certainly willing to tradeoff some breadth or depth of functionality.  However, it's Joel's show.

--Zack

Zack Urlocker
Wednesday, March 26, 2003

I don't think CityDesk has that much market momentum at the moment. There are probably only a few thousand people on the planet that have even heard of it.

Once Fog Creek begin to really market it, that will change rapidly. And they'll want to make sure they really impress all those new users, otherwise they're just selling pipe dreams.

To test, release and support bug fixes, minor enhancements, etc of CityDesk 1.0 would slow down their development of CityDesk 2.0. Delaying that release gives competitors a chance to realise there's a market niche here that's ripe for the picking. Similarly, cutting too much functionality from 2.0 to bring forward the release date would be a let-down.

Fog Creek only has limited resources, so they're probably better off putting all their effort into a killer 2.0 release, even if the wait pisses off a few of us early adopters. If 2.0 is done well, though, we'll forgive them. :-)

Darren Collins
Wednesday, March 26, 2003

just to give this thread a bump.

I do think that CityDesk is starting to  face the Dark force (read, Macromedia).. Contribution does seems eat into CD's market share.

slok
Wednesday, April 02, 2003

*  Recent Topics

*  Fog Creek Home