Grails adoption

classic Classic list List threaded Threaded
121 messages Options
1234567
Reply | Threaded
Open this post in threaded view
|

Re: Grails adoption

Marc Palmer Local
> On learning curve:
>
> Much of the focus of the user guide and tutorials is on Java
> developers. That's one area we could definitely improve on. But what
> are the main areas that non-Java developers find particularly
> difficult?
>

There are two strands here. Documenting stuff better, but also reducing learning curve in the first place by continually smoothing off the rought edges.

> On conferences:
>
> Spring One 2GX, the GR8 conferences, Spring IO in Madrid, the S2G
> Forums, Grails Exchange. So there are some. Spring One 2GX is the
> biggest of those.

The UK Grails Exchange was as we know criminally under PR'd. This can't happen again. Its not just a case of telling people when the next conference is when you're currently at this year's conference.

There are two large (in terms of premium screen real estate) Training panels on grails.org. The first has no clickable links other than an email link. Either way this is unnecessary duplication and not particularly well done :)

Idea:  How about freeing the footer Training panel for one on conferences *past* and future. Link to all the slides and talks for past conferences so people have instant go-to info, and info about what's coming up in the next 12 months.

Marc


---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email


Reply | Threaded
Open this post in threaded view
|

Re: Grails adoption

Sébastien Blanc
It would also be great to know what's SpringSource's position regarding Grails. They have internally a other "similar" product, Spring Roo and has mentioned before in this thread when I search for groovy/grails trainings in Europe it returns an empty list.
Seb


On Thu, Feb 3, 2011 at 12:37 PM, Marc Palmer <[hidden email]> wrote:
> On learning curve:
>
> Much of the focus of the user guide and tutorials is on Java
> developers. That's one area we could definitely improve on. But what
> are the main areas that non-Java developers find particularly
> difficult?
>

There are two strands here. Documenting stuff better, but also reducing learning curve in the first place by continually smoothing off the rought edges.

> On conferences:
>
> Spring One 2GX, the GR8 conferences, Spring IO in Madrid, the S2G
> Forums, Grails Exchange. So there are some. Spring One 2GX is the
> biggest of those.

The UK Grails Exchange was as we know criminally under PR'd. This can't happen again. Its not just a case of telling people when the next conference is when you're currently at this year's conference.

There are two large (in terms of premium screen real estate) Training panels on grails.org. The first has no clickable links other than an email link. Either way this is unnecessary duplication and not particularly well done :)

Idea:  How about freeing the footer Training panel for one on conferences *past* and future. Link to all the slides and talks for past conferences so people have instant go-to info, and info about what's coming up in the next 12 months.

Marc


---------------------------------------------------------------------
To unsubscribe from this list, please visit:

   http://xircles.codehaus.org/manage_email



Reply | Threaded
Open this post in threaded view
|

Re: Grails adoption

johnrellis
I had the same worries but after attending SpringOne in October and 50% of the conference was on Grails and Groovy, these worries were erased....

(The other 50% was pretty much passed out to all other spring technologies)

  It also seemed from the keynotes that Grails was pushing a lot more boundaries than Roo.  

But this is just a lay mans perspective, there's actually Spring people on the list that may be able to give you figures but I thought that was a good metric. 

On Thu, Feb 3, 2011 at 11:51 AM, Sébastien Blanc <[hidden email]> wrote:
It would also be great to know what's SpringSource's position regarding Grails. They have internally a other "similar" product, Spring Roo and has mentioned before in this thread when I search for groovy/grails trainings in Europe it returns an empty list.
Seb


On Thu, Feb 3, 2011 at 12:37 PM, Marc Palmer <[hidden email]> wrote:
> On learning curve:
>
> Much of the focus of the user guide and tutorials is on Java
> developers. That's one area we could definitely improve on. But what
> are the main areas that non-Java developers find particularly
> difficult?
>

There are two strands here. Documenting stuff better, but also reducing learning curve in the first place by continually smoothing off the rought edges.

> On conferences:
>
> Spring One 2GX, the GR8 conferences, Spring IO in Madrid, the S2G
> Forums, Grails Exchange. So there are some. Spring One 2GX is the
> biggest of those.

The UK Grails Exchange was as we know criminally under PR'd. This can't happen again. Its not just a case of telling people when the next conference is when you're currently at this year's conference.

There are two large (in terms of premium screen real estate) Training panels on grails.org. The first has no clickable links other than an email link. Either way this is unnecessary duplication and not particularly well done :)

Idea:  How about freeing the footer Training panel for one on conferences *past* and future. Link to all the slides and talks for past conferences so people have instant go-to info, and info about what's coming up in the next 12 months.

Marc


---------------------------------------------------------------------
To unsubscribe from this list, please visit:

   http://xircles.codehaus.org/manage_email






--
John Rellis
@johnrellis

Reply | Threaded
Open this post in threaded view
|

Re: Grails adoption

Martijn Thieme
In reply to this post by Martijn Vos
The "relative" view of the job postings is even more interesting:

http://www.indeed.com/jobtrends?q=grails%2Cruby%2Cscala%2Cphp%2Cpython%2Cja
va&l=&relative=1


On 2/3/11 10:56 , "Martijn Vos" <[hidden email]> wrote:

>
>
>On 2/2/11 5:37 PM, "Marc Palmer" <[hidden email]> wrote:
>
>>
>>On 2 Feb 2011, at 15:33, Graeme Rocher wrote:
>>
>>> On Wed, Feb 2, 2011 at 1:50 PM, Marc Palmer <[hidden email]> wrote:
>>>>
>>>> On 2 Feb 2011, at 12:36, Mengü Kağan wrote:
>>>>
>>>>> In my opinion Grails is constantly growing. Both in code base and
>>>>>usage. I have met Grails in 2008 and since then I'm watching it very
>>>>>closely and using it for a start-up.
>>>>>
>>>>> 2011/2/2 Daniel Henrique Alves Lima <[hidden email]>
>>>>>      Hi, everybody.
>>>>>
>>>>>      I believe others have asked the following questions before, but
>>>>>i'm
>>>>> looking for new answers:
>>>>>
>>>>>      What do you think about Grails adoption "around the world" (for
>>>>> corporations and private projects)? Is Grails usage growing? Why not?
>>>>> What is holding back Grails?
>>>>>
>>>>>      I'm just wondering...
>>>>>
>>>>
>>>> I have to say that at least sitting here in the Grails shack I'm not
>>>>seeing much in the way of increased job traffic for Grails work. A
>>>>little increase over the last 2-3 years, but not much.
>>>>
>>>> My *completely un-scientific* guy feeling is: Enterprise still won't
>>>>touch Groovy, and startups still go with Rails.
>>>>
>>>> Of course, I have done nothing but Grails for years. I believe the
>>>>user base for Grails is increasing certainly, but I'm not convinced the
>>>>market share growth is stellar in commercial environments yet.
>>>>
>>>> Anybody got some useful stats to prove me wrong?
>>>
>>>
>>>
>>> Well Indeed shows the number of jobs trending in the right direction
>>> http://www.indeed.com/jobtrends?q=grails
>>>
>>> Take from that what you will
>>
>>Not bad, but compare:
>>
>>http://www.indeed.com/jobtrends?q=grails%2Crails%2Cscala%2Cphp&l=
>
>I was hoping that by now, PHP would be losing ground to Ruby, Groovy and
>Python. Or holding it's own. But it just keeps growing.
>
>Here's the graph with all languages rather than framworks:
>
>http://www.indeed.com/jobtrends?q=groovy%2Cruby%2Cscala%2Cphp%2Cpython%2Cj
>a
>va&l=
>
>I added Java too, just to put things in perspective.
>
>
>mcv.
>

Reply | Threaded
Open this post in threaded view
|

Re: Grails adoption

johnrellis
That is interesting... good use of your developer powers!

On Thu, Feb 3, 2011 at 12:39 PM, Martijn Thieme <[hidden email]> wrote:
The "relative" view of the job postings is even more interesting:

<a href="http://www.indeed.com/jobtrends?q=grails%2Cruby%2Cscala%2Cphp%2Cpython%2Cja va&amp;l=&amp;relative=1" target="_blank">http://www.indeed.com/jobtrends?q=grails%2Cruby%2Cscala%2Cphp%2Cpython%2Cja
va&l=&relative=1


On 2/3/11 10:56 , "Martijn Vos" <[hidden email]> wrote:

>
>
>On 2/2/11 5:37 PM, "Marc Palmer" <[hidden email]> wrote:
>
>>
>>On 2 Feb 2011, at 15:33, Graeme Rocher wrote:
>>
>>> On Wed, Feb 2, 2011 at 1:50 PM, Marc Palmer <[hidden email]> wrote:
>>>>
>>>> On 2 Feb 2011, at 12:36, Mengü Kağan wrote:
>>>>
>>>>> In my opinion Grails is constantly growing. Both in code base and
>>>>>usage. I have met Grails in 2008 and since then I'm watching it very
>>>>>closely and using it for a start-up.
>>>>>
>>>>> 2011/2/2 Daniel Henrique Alves Lima <[hidden email]>
>>>>>      Hi, everybody.
>>>>>
>>>>>      I believe others have asked the following questions before, but
>>>>>i'm
>>>>> looking for new answers:
>>>>>
>>>>>      What do you think about Grails adoption "around the world" (for
>>>>> corporations and private projects)? Is Grails usage growing? Why not?
>>>>> What is holding back Grails?
>>>>>
>>>>>      I'm just wondering...
>>>>>
>>>>
>>>> I have to say that at least sitting here in the Grails shack I'm not
>>>>seeing much in the way of increased job traffic for Grails work. A
>>>>little increase over the last 2-3 years, but not much.
>>>>
>>>> My *completely un-scientific* guy feeling is: Enterprise still won't
>>>>touch Groovy, and startups still go with Rails.
>>>>
>>>> Of course, I have done nothing but Grails for years. I believe the
>>>>user base for Grails is increasing certainly, but I'm not convinced the
>>>>market share growth is stellar in commercial environments yet.
>>>>
>>>> Anybody got some useful stats to prove me wrong?
>>>
>>>
>>>
>>> Well Indeed shows the number of jobs trending in the right direction
>>> http://www.indeed.com/jobtrends?q=grails
>>>
>>> Take from that what you will
>>
>>Not bad, but compare:
>>
>>http://www.indeed.com/jobtrends?q=grails%2Crails%2Cscala%2Cphp&l=
>
>I was hoping that by now, PHP would be losing ground to Ruby, Groovy and
>Python. Or holding it's own. But it just keeps growing.
>
>Here's the graph with all languages rather than framworks:
>
>http://www.indeed.com/jobtrends?q=groovy%2Cruby%2Cscala%2Cphp%2Cpython%2Cj
>a
>va&l=
>
>I added Java too, just to put things in perspective.
>
>
>mcv.
>




--
John Rellis
@johnrellis

Reply | Threaded
Open this post in threaded view
|

Re: Grails adoption

Marc Palmer Local

On 3 Feb 2011, at 12:55, John Rellis wrote:

> That is interesting... good use of your developer powers!

LOL. The phrase "not statistically significant" sort of springs to mind.

As does "From small acorns...". But we're, what... 5+ years into the Grails adventure now.

Marc


---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email


Reply | Threaded
Open this post in threaded view
|

Re: Grails adoption

Martijn Vos


On 2/3/11 2:20 PM, "Marc Palmer" <[hidden email]> wrote:

>
>On 3 Feb 2011, at 12:55, John Rellis wrote:
>
>> That is interesting... good use of your developer powers!
>
>LOL. The phrase "not statistically significant" sort of springs to mind.
>
>As does "From small acorns...". But we're, what... 5+ years into the
>Grails adventure now.

Groovy is still one of the youngest languages on the list, though. Python,
Ruby and PHP are (nearly?) as old as Java.
Only Scala is really younger than Groovy.

But yeah, I was also hoping Grails was a bit bigger. The combination of
cool Rubyisms on established Java frameworks should be a really nice
sweetspot.


mcv.


---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email


Reply | Threaded
Open this post in threaded view
|

Re: Grails adoption

Guillaume Laforge-2
On Thu, Feb 3, 2011 at 14:41, Martijn Vos <[hidden email]> wrote:
> [...]
> Only Scala is really younger than Groovy.

Actually, Scala's design started in 2001, with a first release in 2003.
(Groovy was born in 2003 with a first release in 2003)


--
Guillaume Laforge
Groovy Project Manager
Head of Groovy Development at SpringSource
http://www.springsource.com/g2one

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email


Reply | Threaded
Open this post in threaded view
|

Re: Grails adoption

rlovtangen
In reply to this post by Müller, Wolfgang

On Feb 3, 2011, at 1:30 AM, Müller, Wolfgang wrote:


1) It always seems to be a release or two away from being stable.  Maybe we're on the lists too much and see the bugs that will be fixed next time and each time the bugs seem significant enough to give us pause.


I agree with this. Using the latest stable release sometimes *has* big regressions. So far, no-one has worked out a way how to contribute testing in a simple manner that would not mean full participation in the devel cycle, but still give more stability.

Strongly agree on this one. Hoping for a regression free 1.3.7 today ;)

 
5) There are not enough discussion (really any at all) on how to mix this in with an existing Java project, if that's even possible.  It would be great if we could slip some Groovy controllers into our existing Java apps to see how it all works.  Perhaps our more junior programmers or more technical product managers could knock out some demos in Groovy that could be planned to be converted to Java later.  The hope is that our senior programmers would see that it just works and they would decide later they don't have to convert it at all and might even start using it.

This actually is the point where I strongly agree. This is something we're looking into, maybe colleagues of mine can report on that in a couple of months.

This is actually how we started using Grails. We had an existing projects with business logic in regular jars, built with maven. What we needed was a web framework, and Grails was superior to our previous framework of choice, Struts 2.  It was close to trivial to hook Grails up to our business logic in the jars. So we used just the Controller/view/filter parts of Grails, and it worked fine. We used maven for building our Grails frontend as well, so we could still go to project root folder and type "maven install", and get the final war as usual. I think without the maven support, we would not have chosen Grails. Later we have moved the business logic into the Grails module. Most code was just moved untouched into src/groovy, and is slowly getting moved to regular Grails services etc. We also threw out maven, because of the upgrade pain, when almost none of the Grails releases the last year worked fine with maven (except 1.3.4). When we switched to regular Grails build, we got some problems though that we had to work around, because of lack of separation in Grails between the build system and the application. That will hopefully be resolved when Grails switch to Gradle.


Ronny
Reply | Threaded
Open this post in threaded view
|

RE: Grails adoption

Ray Hooker

We have used Grails in two POC projects.  We are a java shop which does most new projects based on Spring, so doing a POC in grails makes a lot of sense.  I can incorporate already built java components seamlessly.  I can also build some specific interface code in Java (e.g., to connect to legacy web services or use specific Java libraries) which can later be used in production.  The architecture and logical design is going to match production.  This is the good part.

 

We have experienced real challenges with the stability of our projects.  There is a great deal of complexity since it is built on Spring. I have done Ruby on Rails and still have code I support.  With Rails, it just works once I get it debugged. Of course Rails is a single stack built from scratch whereas Grails is built on Spring.  With my grails projects, we have had real fits with the database connection pooling and still periodically have trouble with it.  This is an area that is not well described in the grails doc and actually can be real voodoo for Spring developers (i.e., Hibernate tuning).    So that is the challenge.  Perhaps I would work it out if I studied Spring internals in more detail.

 

So we will continue to use Grails as appropriate on POC’s.  I would be hesitant to use it in production code at this point though I haven’t given up on it.

 

Ray

 

From: Ronny Løvtangen [mailto:[hidden email]]
Sent: Thursday, February 03, 2011 9:55 AM
To: [hidden email]
Subject: Re: [grails-user] Grails adoption

 

 

On Feb 3, 2011, at 1:30 AM, Müller, Wolfgang wrote:



1) It always seems to be a release or two away from being stable.  Maybe we're on the lists too much and see the bugs that will be fixed next time and each time the bugs seem significant enough to give us pause.

 

I agree with this. Using the latest stable release sometimes *has* big regressions. So far, no-one has worked out a way how to contribute testing in a simple manner that would not mean full participation in the devel cycle, but still give more stability.

 

Strongly agree on this one. Hoping for a regression free 1.3.7 today ;)



 

5) There are not enough discussion (really any at all) on how to mix this in with an existing Java project, if that's even possible.  It would be great if we could slip some Groovy controllers into our existing Java apps to see how it all works.  Perhaps our more junior programmers or more technical product managers could knock out some demos in Groovy that could be planned to be converted to Java later.  The hope is that our senior programmers would see that it just works and they would decide later they don't have to convert it at all and might even start using it.

This actually is the point where I strongly agree. This is something we're looking into, maybe colleagues of mine can report on that in a couple of months.

 

This is actually how we started using Grails. We had an existing projects with business logic in regular jars, built with maven. What we needed was a web framework, and Grails was superior to our previous framework of choice, Struts 2.  It was close to trivial to hook Grails up to our business logic in the jars. So we used just the Controller/view/filter parts of Grails, and it worked fine. We used maven for building our Grails frontend as well, so we could still go to project root folder and type "maven install", and get the final war as usual. I think without the maven support, we would not have chosen Grails. Later we have moved the business logic into the Grails module. Most code was just moved untouched into src/groovy, and is slowly getting moved to regular Grails services etc. We also threw out maven, because of the upgrade pain, when almost none of the Grails releases the last year worked fine with maven (except 1.3.4). When we switched to regular Grails build, we got some problems though that we had to work around, because of lack of separation in Grails between the build system and the application. That will hopefully be resolved when Grails switch to Gradle.

 

 

Ronny

Reply | Threaded
Open this post in threaded view
|

Re: Grails adoption

bdrhoa
In reply to this post by John Thompson
That sounds like the perfect tutorial app. It's more of a business application than a wiki like grails.org is.

I'll just add, the thing I think is most lacking is easy support for one-to-many and many-to-many applications. In particular, if the the scafolding would generate a (jquery/AJAJ) 1 screen form for one-to-many relationships, that would make scaffolding an 80-90% solution vs the 50-60% solution it is now.
---------------------------
www.maf.org/rhoads
www.ontherhoads.org


On Thu, Feb 3, 2011 at 2:52 AM, John Thompson <[hidden email]> wrote:

Grailscasts.com interesting thought.

I have a small app I was thinking about setting free to the open source
world.  Its a time and expense tracker - nothing too sexy.  I have under 80
hrs invested in it.

But the little app is using Spring Security Core, Spring Security UI, jQuery
UI, Blueprint CSS, and I even mixed in some ExtJS components at the end.

I was thinking it might be fun to rip it apart in a blog series or some web
casts for beginners.

Thoughts?

-----
JT
jts-blog.com
--
View this message in context: http://grails.1312388.n4.nabble.com/Grails-adoption-tp3253881p3257717.html
Sent from the Grails - user mailing list archive at Nabble.com.

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

   http://xircles.codehaus.org/manage_email



Reply | Threaded
Open this post in threaded view
|

Re: Grails adoption

Peter Bell-5
In reply to this post by Scott6666

On Feb 2, 2011, at 8:04 PM, Scott Eisenberg wrote:
You might get fired if you try Ruby On Rails and it doesn't scale.  You should given what Twitter went through (unless Rails 3 is now very different because of the Twitter experience, I don't know).  

It's a bit OT for the list, but I don't think Twitter had problems because of the stack they used. They had problems because they were impossibly successful, needed to scale incredibly quickly and seemed to believe at the start that scaling was just a matter of adding boxes. Most startups would blow up if they had that kind of ramp up and minimal experience at scaling web apps.

That's why I'm not sure what all those start-ups are doing ROR.  Except a VC does not fire you for being trendy; makes it easier to sell you off to someone.

They do it because it really is very productive and capable. And VC's would absolutely fire you for being trendy. Have you ever done a raise for a start up? I founded and raised for a dot-com in 2000 and while many mistakes were made, I never noticed any VC's asking me to be trendy. They wanted me to *make* *them* *money*.
 
How many apps have you written in Rails? I split my time evenly between Rails and Grails (with a little Scala on the side and Java when I really must). Grails is an excellent platform and I'm a huge fan of all of the gr8 technologies and present on them frequently. The platform still isn't as mature in areas as Rails. We still don't have a comparable stack to Cucumber, Capybara, Factory Girl, Rspec, Fakeweb and vcr. When you see how easy it is to write really good tests - even on API's in Rails it makes you realize that even the amazing work Peter and Luke are doing on Spock and Geb is still not replicating what exists in the Rails world. We are behind the curve in testing and it's something we need to work on (I'm presenting on Spock and Geb this year and hope to start contributing if I ever get a spare moment).

For a Java team/shop, Groovy and Grails are a no brainer for increasing productivity. Outside that community I think Grails has more of a fight on its hands just due to the comprehensiveness of the gems available to Ruby and Rails developers. 

Best Wishes,
Peter
 


PS I truly applaud Graeme for personally helping on the list.  I do think that he's too close to it to understand what it all looks like to a newbie now.



On Feb 2, 2011, at 7:30 PM, Müller, Wolfgang wrote:

While I have that confidence, examples would be nice. Nosy question: Do you get such examples from the PHP community?
 
This would indeed be interesting. Again the question: Are ruby/PHP endgames out there?



Reply | Threaded
Open this post in threaded view
|

Re: Grails adoption

Peter Bell-5
In reply to this post by John Thompson
I think that would be amazing and very valuable. More great example projects would be a really useful thing  especially using spring security core and UI and working with blueprint and jQuery. Please announce on this list if you publish it!

Best Wishes,
Peter

On Feb 3, 2011, at 4:52 AM, John Thompson wrote:

>
> Grailscasts.com interesting thought.
>
> I have a small app I was thinking about setting free to the open source
> world.  Its a time and expense tracker - nothing too sexy.  I have under 80
> hrs invested in it.
>
> But the little app is using Spring Security Core, Spring Security UI, jQuery
> UI, Blueprint CSS, and I even mixed in some ExtJS components at the end.
>
> I was thinking it might be fun to rip it apart in a blog series or some web
> casts for beginners.
>
> Thoughts?
>
> -----
> JT
> jts-blog.com
> --
> View this message in context: http://grails.1312388.n4.nabble.com/Grails-adoption-tp3253881p3257717.html
> Sent from the Grails - user mailing list archive at Nabble.com.
>
> ---------------------------------------------------------------------
> To unsubscribe from this list, please visit:
>
>    http://xircles.codehaus.org/manage_email
>
>


---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email


Reply | Threaded
Open this post in threaded view
|

Re: Grails adoption

Peter Bell-5
In reply to this post by John Thompson
But not as easy as "git push heruko master". I'm with Tomas. I'd think there would be a business opportunity there for someone . . .

Best Wishes,
Peter

On Feb 3, 2011, at 5:31 AM, John Thompson wrote:

I deployed a grails app on EC2 via elastic beanstalk this weekend.  *very* cool.  And very easy.

On Thu, Feb 3, 2011 at 5:27 AM, tomas lin [via Grails] <<a href="x-msg://758/user/SendEmail.jtp?type=node&amp;node=3257789&amp;i=0" target="_top" rel="nofollow">[hidden email]> wrote:
Have you tried finding a place to host your toy grails apps?

In Rails, I push button deploy to Heroku. In Django, they're getting Djangy.

In Grails, we either have to write our apps for the App Engine or
spend a ton of time getting Amazon set up since Cloud Foundry does not
support the micro instance.

I think there would be more visibility / enthusiasm for grails if
there was a freemium cloud option that did not require spending two or
three whole days setting up a linux VM to do.

On Wed, Feb 2, 2011 at 12:09 PM, Daniel Henrique Alves Lima
<[hidden email]> wrote:

>      Hi, everybody.
>
>      I believe others have asked the following questions before, but i'm
> looking for new answers:
>
>      What do you think about Grails adoption "around the world" (for
> corporations and private projects)? Is Grails usage growing? Why not?
> What is holding back Grails?
>
>      I'm just wondering...
>
>      Best regards,
>
>              Daniel.
>
> ---------------------------------------------------------------------
> To unsubscribe from this list, please visit:
>
>   http://xircles.codehaus.org/manage_email
>
>
>
---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email





If you reply to this email, your message will be added to the discussion below:
http://grails.1312388.n4.nabble.com/Grails-adoption-tp3253881p3257785.html
To unsubscribe from Grails adoption, click here.



View this message in context: Re: Grails adoption
Sent from the Grails - user mailing list archive at Nabble.com.

Reply | Threaded
Open this post in threaded view
|

Re: Grails adoption

Peter Bell-5
In reply to this post by pledbrook

On Feb 3, 2011, at 6:21 AM, Peter Ledbrook wrote:

> On learning curve:
> Much of the focus of the user guide and tutorials is on Java
> developers. That's one area we could definitely improve on. But what
> are the main areas that non-Java developers find particularly
> difficult?

Well, if they're coming from PHP they are hosed. Not knowing about typing means type errors will completely baffle them. They won't know about jars, won't have experience with war files or jvm tuning, and they will know nothing about dependency injection, spring, spring security, or hibernate. They may not even have used an ORM before.

I'd say the focus on Java devs or at least devs who know some Java is reasonable. I guess the quick win would be better links to spring documentation for J2EE Java devs who don't have a background using Spring. I think to take a PHP or Perl programmer and get them up to speed might be too much of a project. Not saying they can't do it, but I'm not convinced it's our core market - especially given ho many Java developers there are in the world.

Just my 2c

Best Wishes,
Peter


---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email


Reply | Threaded
Open this post in threaded view
|

Re: Grails adoption

fabien7474
In reply to this post by tomas lin
+1. Also Hosting costs are higher with Grails that with Rails or PHP.

tomas lin wrote
Have you tried finding a place to host your toy grails apps?

In Rails, I push button deploy to Heroku. In Django, they're getting Djangy.

In Grails, we either have to write our apps for the App Engine or
spend a ton of time getting Amazon set up since Cloud Foundry does not
support the micro instance.

I think there would be more visibility / enthusiasm for grails if
there was a freemium cloud option that did not require spending two or
three whole days setting up a linux VM to do.
Reply | Threaded
Open this post in threaded view
|

Re: Grails adoption

Joshua Kehn
Really? What makes you say that?

Aside from shared hosting, which I don't consider really hosting, the costs are relatively the same. 

Regards,

-Josh
____________________________________
Joshua Kehn | [hidden email]
http://joshuakehn.com

On Feb 3, 2011, at 11:32 AM, fabien7474 wrote:


+1. Also Hosting costs are higher with Grails that with Rails or PHP.


tomas lin wrote:

Have you tried finding a place to host your toy grails apps?

In Rails, I push button deploy to Heroku. In Django, they're getting
Djangy.

In Grails, we either have to write our apps for the App Engine or
spend a ton of time getting Amazon set up since Cloud Foundry does not
support the micro instance.

I think there would be more visibility / enthusiasm for grails if
there was a freemium cloud option that did not require spending two or
three whole days setting up a linux VM to do.



--
View this message in context: http://grails.1312388.n4.nabble.com/Grails-adoption-tp3253881p3258445.html
Sent from the Grails - user mailing list archive at Nabble.com.

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

   http://xircles.codehaus.org/manage_email



Reply | Threaded
Open this post in threaded view
|

Re: Grails adoption

xmly
In reply to this post by fabien7474
I think amazon elastic beanstalk is already there...


On Thu, Feb 3, 2011 at 11:32 AM, fabien7474 <[hidden email]> wrote:

+1. Also Hosting costs are higher with Grails that with Rails or PHP.


tomas lin wrote:
>
> Have you tried finding a place to host your toy grails apps?
>
> In Rails, I push button deploy to Heroku. In Django, they're getting
> Djangy.
>
> In Grails, we either have to write our apps for the App Engine or
> spend a ton of time getting Amazon set up since Cloud Foundry does not
> support the micro instance.
>
> I think there would be more visibility / enthusiasm for grails if
> there was a freemium cloud option that did not require spending two or
> three whole days setting up a linux VM to do.
>
>

--
View this message in context: http://grails.1312388.n4.nabble.com/Grails-adoption-tp3253881p3258445.html
Sent from the Grails - user mailing list archive at Nabble.com.

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

   http://xircles.codehaus.org/manage_email



Reply | Threaded
Open this post in threaded view
|

Re: Grails adoption

Björn Wilmsmann-2
In reply to this post by Daniel Henrique Alves Lima
Hi there,

in my opinion the problem is that Grails essentially comes from an 'enterprise' background. This background comes with a lot of cultural baggage and implications.

While Grails certainly is a huge improvement over other Java 'enterprise' frameworks like JSF or Struts it still falls short of matching Rails, which is kind of like the gold standard for web development framework nowadays.

Now, while tying in with existing Java frameworks and 'enterprise' requirements isn't a bad thing, a startup or SME starting a new web app couldn't care less. They'll more often than not choose a tool that gets the job done as quickly and elegantly as possible. From my point of view, Grails still doesn't perform as well as Rails in this respect. To name but a few important issues where Rails easily outperforms Grails:

1.) Startup time: Waiting like 10-30s or more for a web server to start or unit tests to execute is just ridiculous nowadays.
2.) Error messages: Java / Groovy / Grails stack traces are hideous. I don't care about abstract underlying exceptions, Hibernate proxy-somethings and Spring whatnots. When developing web apps I just want to see the exact line in my code that caused the error. Sure, things aren't always as simple as this but this behaviour should be the general case.
2.) Unit testing: In my opinion, unit tests with Grails still require too much boilerplate code.
3.) Helpers: Rails offers much more and richer helper classes that help you accomplishing common tasks in web app development. Yes, for each of these there is a Java equivalent but often these Java equivalents are far from easy to use and require lots of useless code just to get them working in common cases.
4.) Plugins: There are much more and especially more stable plugins for Rails.
5.) Leaky abstractions: This mostly concerns Spring- and GORM-related stuff. Once you run into something that can't be accomplished with Grails right out of the box you have to deal with Hibernate internals. ActiveRecord isn't perfect either but at least you can revert to vanilla SQL, whose behaviour often is much more predictably than HQL.

Certainly, startups arent everything. Most well-paid jobs - especially for Java developers - are in the 'enterprise' segment. However, most of those are far from what's usually considered fun or exciting. Let's face it, crunching numbers for a bank or printing contracts for an insurance company just isn't as sexy as the latest Web 2.0 app.

Apart from hype and 'sexiness', Grails also doesn't seem to have a lot of support in the 'enterprise' crowd. Some people tend to become very religious about static typing (and why it's the one-size-fits-all solution for any problem they can think of) and the biggest non-issue of all time: Scalability (If you're Twitter you have scaling problem, if you're Google you have a scaling problem. However, if you're an insurance company with 10,000 employees chances are you don't have a scaling problem. Even if you do, you should probably worry a lot more about your crappy WebSphere and SAP stuff than about your web development framework).

On 2 February 2011 13:09, Daniel Henrique Alves Limabld <[hidden email]> wrote:

>      Hi, everybody.
>
>      I believe others have asked the following questions before, but i'm
> looking for new answers:
>
>      What do you think about Grails adoption "around the world" (for
> corporations and private projects)? Is Grails usage growing? Why not?
> What is holding back Grails?
>
>      I'm just wondering...
>
>      Best regards,
>
>              Daniel.
>
> ---------------------------------------------------------------------
> To unsubscribe from this list, please visit:
>
>   http://xircles.codehaus.org/manage_email
>
>
>



--
Viele Grüße / Best regards, Björn Wilmsmann
Geschäftsführer / CEO

----------------------
MetaSieve GmbH

Universitätsstr. 142
D-44799 Bochum
Germany

Phone: +49-(0)234-7089300
Mobile: +49-(0)151-25209060
Fax: +49-(0)30-46999-1267

E-mail: [hidden email]
http://www.metasieve.com/

Amtsgericht Bochum, HRB 12288
----------------------




---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email


Reply | Threaded
Open this post in threaded view
|

Re: Grails adoption

Marc Palmer Local

On 3 Feb 2011, at 17:02, Björn Wilmsmann wrote:

> Hi there,
>
> in my opinion the problem is that Grails essentially comes from an 'enterprise' background. This background comes with a lot of cultural baggage and implications.
>
> While Grails certainly is a huge improvement over other Java 'enterprise' frameworks like JSF or Struts it still falls short of matching Rails, which is kind of like the gold standard for web development framework nowadays.
>
> Now, while tying in with existing Java frameworks and 'enterprise' requirements isn't a bad thing, a startup or SME starting a new web app couldn't care less. They'll more often than not choose a tool that gets the job done as quickly and elegantly as possible. From my point of view, Grails still doesn't perform as well as Rails in this respect. To name but a few important issues where Rails easily outperforms Grails:
>
> 1.) Startup time: Waiting like 10-30s or more for a web server to start or unit tests to execute is just ridiculous nowadays.
> 2.) Error messages: Java / Groovy / Grails stack traces are hideous. I don't care about abstract underlying exceptions, Hibernate proxy-somethings and Spring whatnots. When developing web apps I just want to see the exact line in my code that caused the error. Sure, things aren't always as simple as this but this behaviour should be the general case.

Several people have mentioned this. Is the stacktrace sanitizer not working for you? If not then one of the following is happening:

* Your full stack trace log may be configured to dump to stdout (you should see sanitized + unsanitized in this case though)
* There are some packages that should be sanitized but aren't for grails apps
* There is a missing sanitize() call in Grails somewhere

This can be fixed for any of the three above, if we know what it is.

Also I would like to see some runtime config / plugin hooks for adding to the list of packages to filter. Better still, we could probably make the sanitizing plugin-aware by checking to see which plugin declared the class that is in the trace, and:

a. filter out all stack entries from code provided by plugins (from the sanitized version only, obviously)
b. detect if the exception actually originated inside a plugin with no app code above it - and show the plugin trace entries in that case
c. invert the whole process and have config for your app's base package, and have sanitize filter EVERYTHING except stuff in that package

The stacktrace sanitizer is relatively straightforward so this would be pretty easy to do. The other area for improvement is to see if we can make it so we don't have to remember to call sanitize or deepSanitize. Perhaps by applying some special handling at a higher level e.g. ThreadGroup.

> 2.) Unit testing: In my opinion, unit tests with Grails still require too much boilerplate code.
> 3.) Helpers: Rails offers much more and richer helper classes that help you accomplishing common tasks in web app development. Yes, for each of these there is a Java equivalent but often these Java equivalents are far from easy to use and require lots of useless code just to get them working in common cases.
> 4.) Plugins: There are much more and especially more stable plugins for Rails.
> 5.) Leaky abstractions: This mostly concerns Spring- and GORM-related stuff. Once you run into something that can't be accomplished with Grails right out of the box you have to deal with Hibernate internals. ActiveRecord isn't perfect either but at least you can revert to vanilla SQL, whose behaviour often is much more predictably than HQL.
>

I 100% agree with 5. The major attraction of GORM is that it hides Hibernate. Except that you find later that it doesn't. I've always argued that GORM should be described 100% in its own terms and behaviours, so there is a single reference document for all your experience.

Hopefully we are moving in that direction now there is the GORM TCK and a plethora of back end implementations. At least the documentation can be focussed more i.e. gorm-hibernate docs can cover all the hibernate you need to know to use the plugin, without polluting the primary GORM docs.

> Certainly, startups arent everything. Most well-paid jobs - especially for Java developers - are in the 'enterprise' segment. However, most of those are far from what's usually considered fun or exciting. Let's face it, crunching numbers for a bank or printing contracts for an insurance company just isn't as sexy as the latest Web 2.0 app.
>
> Apart from hype and 'sexiness', Grails also doesn't seem to have a lot of support in the 'enterprise' crowd. Some people tend to become very religious about static typing (and why it's the one-size-fits-all solution for any problem they can think of) and the biggest non-issue of all time: Scalability (If you're Twitter you have scaling problem, if you're Google you have a scaling problem. However, if you're an insurance company with 10,000 employees chances are you don't have a scaling problem. Even if you do, you should probably worry a lot more about your crappy WebSphere and SAP stuff than about your web development framework).
>

I am with you 100% on this.

But you can't really blame people for taking the easy reactionary option. There are high level issues as mentioned in this thread.

It is also notoriously hard, if not impossible, to gain "victory" over the first past the post (Rails) especially if that product continues to innovate.

Marc


---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email


1234567