Re: Kickstarter was not successful... but it did help things...

[Doc O'Leary]

In article <mailman.10196.1387835000.10748.discuss-gnustep@gnu.org>,
 Gregory Casamento <greg.casamento@gmail.com> wrote:

> I am not trying to refute your points, I¹m trying to discuss them.  You seem 
> very adamant on banging on how wrong we are without being willing to discuss 
> your points rationally.

As always, just the opposite.  I quite welcome rational discussions.  I 
don't think I've once called them hot air or chided someone for not 
being positive enough.

> I don¹t know where the idea that they are post-hoc (i.e. looking at the data 
> and making a determination after the fact) comes from.   These are reasons 
> which have been told to me personally by users I¹ve spoken to at conferences 
> and such as to why they themselves didn¹t adopt GNUstep.

That doesn't mean they weren't justifying their decision after the fact.  
If you actually wanted to get experimental about it, you would set 
people up with a level playing field at the start and see why they 
*actually* decided to go one way or another.  Most often it is a matter 
of politics or business agreements rather than anything fundamental 
about a technology itself.

This even applies to the boom of ObjC for Apple.  Did some technical 
hurdle get overcome that had been holding it back?  Nope, just a bunch 
of people who realized there was money to be made in the App Store and 
they'd do whatever it took to chase those riches.

> I agreed with you so many times on the point that we don¹t communicate our 
> message effectively it¹s not even funny.   Next point please.

No.  Don't move on from that when it matters *far* more than code at 
this point.  Resolve to *fix* it and we could be progressing in a whole 
different direction.

> >> 1) It has an outdated look
> 
> This was actually the PRIMARY reason and the most common reason why I¹m told 
> people don¹t want to use it.  To most people our GUI is just not attractive. 
> Sorry for anyone that might piss off on the project, but it is true and it¹s 
> been repeated for years.

Still: doesn't matter.  It's a lie masking some other issue.  I say that 
because you know the truth: the GNUstep API is intended to be platform 
independent.  No developer worth talking to is going to abandon the 
frameworks because *one* platform doesn't have a particular look that is 
to their liking.  The same code could run on a shiny new Mac; how the 
hell is that outdated?

Without being able to question them, it's hard to say what they were 
actually justifying.  Perhaps it was that they wanted better support for 
their preferred platform.  Whatever the root issue is, you have to get 
at it before you can actually deliver them a message that brings them 
onboard.

> >> 2) Most developers in Open Source don1t like Objective-C.  They consider 
> >> it 
> >> UGLY.
> 
> Again.  This is something developers have told me again and again.  They say 
> that Objective-C has an ³ugly syntax.²   Most developers, in my experience, 
> would rather use GTK or Qt on the Mac and write applications simply because 
> it uses C++.

C++ is ugly to the bone.  What I *actually* hear you saying is that some 
developers are incapable of learning new things, even if the actual 
utility to them is greater.  I, too, have heard some developers complain 
about verbose method names or other surface issues that not only aren't 
ugly, just different, but actually *help* you make sense of the code in 
the long term.  Looking at those code those kind of developers crank 
out, I'd be happy to not work with them.

And, again, the whole issue is clearly moot when you look at the bigger 
picture.  If ObjC were *actually* too ugly, Apple wouldn't even be 
getting any takers for doing iOS apps.  Pure justification by whoever it 
is you were talking to.

> >> 3) Many people assume GNUstep is an implementation of OpenStep only, 
> >> mainly 
> >> because of it1s look and feel and they don1t go beyond that point to try 
> >> to 
> >> discover more.
> > 
> > That's on GNUstep for failing to better communicate.
> 
> It¹s stated right on our front page in plain English that we are an 
> implementation of Cocoa.   It¹s also something I¹ve repeated again and again 
> at conferences, on shows and on my blog.  If you can tell me a better way to 
> tell people, please let me know.

What's so hard to understand?  Nobody who isn't in the inner circle 
knows what "Cocoa" means.  Just like most people who use Cocoa don't 
know what NEXTSTEP was, even when it gets stuck at the front of all 
their classes.  I can't tell you I'm sure what the right message for 
this would be at this point, though.  That what I'd been hoping would be 
the direction of these discussions.

My high-level stab at it was the slogan.  To me, GNUstep represents a 
way to take the technology that has made Apple such a success and open 
it up to developers on every platform.  Would your business enjoy that 
kind of success?  Let me tell you more . . .  Is that not a message they 
would respond to?  There's no sort of message like that on the web site.

> >> 4) GNUstep hasn1t implemented iOS/CocoaTouch APIs.
> > 
> > That certainly hasn't helped for the last 5 years, but what about before 
> > that?  What about *any* serious efforts to port Mac software?  Or ports 
> > *to* the Mac of GNUstep software?
> 
> There are a number of apps which have been ported to Windows and Linux from 
> the Mac.  Eggplant being one of them.  There are more on our ³GNUstep 
> successes² page on the wiki. 
> 
> http://wiki.gnustep.org/index.php/Success_Stories 

Woof!  Talk about ugly.  Another tangent suggestion: get promotional 
material out of the wiki.

But, no, what I'm talking about is implementing some internal 
initiatives to highlight GNUstep.  For example, there are tons of open 
source apps that have a Mac version (either as a port or the primary 
platform).  Set up an effort to "dogfood" them over to GNUstep yourself.  
One a week might be too much, but how about an "App of the Month" effort 
to publicize what GNUstep offers?  Even if it doesn't always work, the 
aim is to let people know things are always happening.  My nomination 
for the first one is a doozy: VLC.

> >> 5) in the free software world people were too concentrated on beating 
> >> Apple 
> >> and Microsoft to even think about implementing an API which was defined by 
> >> NeXT/Apple.
> 
> The most common response I get from people is that they didn¹t know that 
> GNUstep existed.  This goes to the point of our message.  

And the main reason I say what you need most is to seriously address 
*that* rather than code.

> >> With the collapse of GNOME due to GNOME31s failures, the above may change 
> >> and 
> >> now, indeed, might be the time for GNUstep to riseS but we shall see.
> > 
> > There's no real need to wait and see.  GNUstep will continue to flounder 
> > so long as it fails to deliver a message that people want to hear.  That 
> > *should* be the takeaway from the dismal crowd funding effort.  Somehow 
> > I'm a jerk for pointing this out.
> 
> I agree that our message is unclear and that it needs updating.  

Then resolve to do something about it.

> Also, Please don¹t put words in my mouth, I never said you were a jerk for 
> pointing it out.  I said you were being a jerk WHILE pointing it out, there¹s 
> a difference. :)

Fair enough.  :-)

> The ³dismal² crowd funding effort failed because of two very basic reasons:
> 
> 1) Set the goals too high and the timeframe too short
> 2) What I proposed was not really clear

Nope.  You simply didn't tell people what they wanted to hear.  Nobody 
cares about 10.6 anymore, nor does the average user really know that 
that equates to.  WebKit also has no inherent meaning and, in reality, 
is not a core part of many existing apps.  It'd be nice to have to round 
things out, sure, but not *that* critical.  And you got nowhere close to 
fooling anyone that they'd be able to run Mac binaries on Linux any time 
soon.

What people actually want are "stupid" thing.  You probably would have 
shot through the roof if you just said your aim was to bring Angry Birds 
to every computer.  But, as I've said before, even a more reasonable 
goal like UIKit would have been better received.  The external message 
of porting iOS apps is much more clear.

-- 
iPhone apps that matter:    http://appstore.subsume.com/
My personal UDP list: 127.0.0.1, localhost, googlegroups.com, theremailer.net,
    and probably your server, too.