Re: Upcoming 2.0 of LedgerSMB

["Adam Thompson" <..hidden..>]

> From: David F. Skoll [mailto:..hidden..]
>
> Joshua D. Drake wrote:
> > Spoken like a true geek. This isn't hard folks. The web has the whole
> > print thing down as long as you are using a modern browser. Simple
> > web-pdf tools exist as well.
> 
> Color me skeptical.  There a reason most small-business accounting
> systems are *not* web-based.  It's a real PITA to get consistent good-
> quality print output via the Web without using PDF-generation tools that
> completely bypass HTML+CSS.

I'm not sure I agree that LaTeX is the ideal PDF-generation tool, but I do agree with David about the quality of web-based printing.  It generally sucks.  
I just can't agree with "the web has the whole print thing down" - I can't reliably print high-quality output from the web.  IE - the less said the better.  FireFox is OK, but it's still a pain to produce PDF-quality output.  Google Chrome is surprisingly bad at printed output on any platform I've tried so far: the kerning is either too aggressive or improperly implemented, I can't tell which.  
PDF definitely isn't perfect, but it is at least a useful lowest-common-denominator.

> >> Realistically, installing and customizing LedgerSMB is currently done
> >> by Linux and FOSS geeks, and I don't see that changing any time soon.
> > You just made my point. One of my goals for 2.0 is to change that or
> > at least make large leaps toward it.
> Again, I don't mean to be discouraging, but I doubt that's the last
> remaining hurdle for most Quickbooks or Peachtree users.  LedgerSMB
> has far more serious problems to worry about (eg, the Ontario HST
> nightmare) than template configuration.

Echoing David's comments, I would never attempt to position LSMB against Quickbooks or Simply.  (Or Peachtree, although it's pretty rare on the ground here anyway.)

My biggest problem with LSMB right now - and the reason I'm not using it right now - is the installation is, um, obscure, even by FOSS standards.  I found it impossible to get 1.3 up and running by following the documentation; there appeared to be significant gaps where devs & experienced users "just knew" what to do.  (The situation for <1.3 wasn't nearly as bad, but I didn't want to run the old code base.)

There is a significant untapped market for FOSS (and FOSS-like) supporters, as well as anyone who needs a multi-user accounting solution that doesn't require an additional $10k in hw+sw costs.  That market will remain untapped as long as the barrier to entry is as high as it is right now.

My $0.02 is that the installation needs to be worked on first - I'm not going to even get to the point of complaining about features that don't work properly until the installation is reasonably-well documented.  Note that it doesn't have to be *easy* or *simple*, but it either has to be easy and simple *or* well-documented.  In fact, scratch "well"-documented, following the docs just has to actually work.

Regarding features that don't work... in my opinion, features that aren't fully implemented should be either yanked from 2.0 or hidden from view.  Timesheets would be an excellent example of that - the stubs appear to all be there but it's not quite finished, with no indication (other than the mailing list) that it's not finished.

I gave up on using LSMB in the short term simply because it isn't quite polished enough, and I don't want to spend lots of time getting up and running with an accounting system.  Administrative overhead, if you will... the same reason I'll be ditching Gentoo on my primary file/print/mail server this week and switching to either CentOS or Ubuntu LTS.  I don't mind filing bug reports and the occasional patch, but I'm not into heavy lifting just so I can print a handful of invoices per month...

Anyway, that's my two cents' worth.
-Adam Thompson
 ..hidden..