What is highest priority for Scala to succeed in corporate world (Should be in scala-debate?) ?

I think your list is exactly right; if anything I'd just emphasize the first two points.

On Sat, Nov 5, 2011 at 5:51 PM, Ken McDonald <ykkenmcd@gmail.com> wrote:

Even though this perhaps should be in scala-debate, I'm posting it here because:    1) It's really very important.     2) I've never seen a similar question posted.
This comes after reading a few of the "White elefant" posts.
The question is simple: List, in descending order of priority, what you feel needs to be addressed/fixed/whatever for Scala to succeed in the corporate (and hence IMHO ultimately the "real") world.
My list:
1. IDE Support. Scala must have rock-hard support in at least one IDE, and must have "pretty good" support in Eclipse, since it's the de facto standard. I don't think we're close yet. This alone could lose Scala the language wars.

I think Scala must have rock-hard support in _most major IDEs_.  This includes both Eclipse and NetBeans; IDEA is also a good candidate since it seems pretty advanced already.  If I had to switch my IDE as _well_ as my language, that would be an additional barrier, I'd think.
 

2. Documentation. Scaladoc documentation needs to be expanded on greatly; at the same time Scaladoc itself could stand some enhancements, though exactly where is less obvious. I am complicit in this; I keep on intending to devote some serious time to enhancing scaladoc, and keep failing to do so, dangit!

Indeed.  Scala's core API should be as well documented as Java's.  I rarely find that the Java documentation says far more than I need, but I often find that it just barely says enough for me to do something useful.  In Scala, it's often several hours of poking around, looking at source code, blog posts, and StackOverflow, before I figure out something nontrivial that I haven't used before.

  --Rex

all my votes go to "ide support". and a little tiny enhancement to the jvm: i want to be able to write List(1,2,3).filter(_ == 2) in my debugger

never had a problem with documentation. maybe i have magic eyes or i am a google prime customer, but i never have to search long until i find an answer.

Am 05.11.2011 22:58, schrieb Rex Kerr:

CAP_xLa1DukTYMcZWwHqC3f6VORZiOW0M-mJ+JO_fdCmc3mb55A [at] mail [dot] gmail [dot] com" type="cite">I think your list is exactly right; if anything I'd just emphasize the first two points.

On Sat, Nov 5, 2011 at 5:51 PM, Ken McDonald <ykkenmcd [at] gmail [dot] com" rel="nofollow">ykkenmcd@gmail.com> wrote:

Even though this perhaps should be in scala-debate, I'm posting it here because:     1) It's really very important.     2) I've never seen a similar question posted.
This comes after reading a few of the "White elefant" posts.
The question is simple: List, in descending order of priority, what you feel needs to be addressed/fixed/whatever for Scala to succeed in the corporate (and hence IMHO ultimately the "real") world.
My list:
1. IDE Support. Scala must have rock-hard support in at least one IDE, and must have "pretty good" support in Eclipse, since it's the de facto standard. I don't think we're close yet. This alone could lose Scala the language wars.

I think Scala must have rock-hard support in _most major IDEs_.  This includes both Eclipse and NetBeans; IDEA is also a good candidate since it seems pretty advanced already.  If I had to switch my IDE as _well_ as my language, that would be an additional barrier, I'd think.
 

2. Documentation. Scaladoc documentation needs to be expanded on greatly; at the same time Scaladoc itself could stand some enhancements, though exactly where is less obvious. I am complicit in this; I keep on intending to devote some serious time to enhancing scaladoc, and keep failing to do so, dangit!

Indeed.  Scala's core API should be as well documented as Java's.  I rarely find that the Java documentation says far more than I need, but I often find that it just barely says enough for me to do something useful.  In Scala, it's often several hours of poking around, looking at source code, blog posts, and StackOverflow, before I figure out something nontrivial that I haven't used before.

  --Rex

I agree with your list. I'd probably add making the core libs (particularly thinking of collections) faster and lower-memory/burn. This probably includes making specialization work (better?) and aggressively inlining away typeclasses. When performance really does matter there is no substitute for fast code. If the scaladocs are beefed up, I'd probably place performance at 3 in place of the cookbook.

On 5 November 2011 21:51, Ken McDonald <ykkenmcd@gmail.com> wrote:

Even though this perhaps should be in scala-debate, I'm posting it here because:    1) It's really very important.    2) I've never seen a similar question posted.
This comes after reading a few of the "White elefant" posts.
The question is simple: List, in descending order of priority, what you feel needs to be addressed/fixed/whatever for Scala to succeed in the corporate (and hence IMHO ultimately the "real") world.
My list:
1. IDE Support. Scala must have rock-hard support in at least one IDE, and must have "pretty good" support in Eclipse, since it's the de facto standard. I don't think we're close yet. This alone could lose Scala the language wars.
2. Documentation. Scaladoc documentation needs to be expanded on greatly; at the same time Scaladoc itself could stand some enhancements, though exactly where is less obvious. I am complicit in this; I keep on intending to devote some serious time to enhancing scaladoc, and keep failing to do so, dangit!
3. A "Scala Cookbook". I'm amazed one isn't out or on the web already. I know, there's lots on stackoverflow and other sites but it's not the same. Scala is such a neat language, that a newcomer could easily get caught up and read for hours on a good cookbook, saying things like, "Wow, I can do it in just one line...".

And your suggestions? Recommended three or less because probably no one will read past the third one :-).


Ken

--
Dr Matthew PocockIntegrative Bioinformatics Group, School of Computing Science, Newcastle University mailto: turingatemyhamster@gmail.comgchat: turingatemyhamster@gmail.com msn: matthew_pocock@yahoo.co.ukirc.freenode.net: drdozerskype: matthew.pocock tel: (0191) 2566550mob: +447535664143

Hi,

Ken McDonald wrote:
> 1. IDE Support. Scala must have rock-hard support in at least one IDE, and must have "pretty good" support in Eclipse, since it's the de facto standard. I don't think we're close yet. This alone could lose Scala the language wars.

What other JVM language (Java aside) has better IDE support than Scala?

(Also, I find "language wars" a strange term.)

Kind regards
Andreas

I think you need to define what "rock-hard support" in an IDE means.  For example, the Eclipse plugin keeps on moving forward in terms of functionality, but where exactly is the version of it for Scala 2.9.1?  It also consumes a ton of memory and tends to "just sit there" a lot.  It's basically unusable in my work environment, while the version quite a while back, while buggier and with far fewer features, was usable.  I think the NetBeans plugin also took a step sideways when it integrated with the presentation compiler.  In my opinion the only really usable IDE for Scala right now is Emacs+ENSIME.  Maybe the ENSIME integrations for Vim and JEdit work well, too.  I'm fine with Emacs, but I wouldn't try to force in on anyone.

The question is simple: List, in descending order of priority, what you feel needs to be addressed/fixed/whatever for Scala to succeed in the corporate (and hence IMHO ultimately the "real") world.

1) The speed of the compiler. --
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Brian Schlining
bschlining@gmail.com

That basically. 

On 6 Nov 2011, at 01:52, "Brian Schlining" <bschlining@gmail.com> wrote:

The question is simple: List, in descending order of priority, what you feel needs to be addressed/fixed/whatever for Scala to succeed in the corporate (and hence IMHO ultimately the "real") world.

1) The speed of the compiler.  

When you talk about real world, it is important that some of those very nice features of the language which are noticably slower than what a novice would expect should be de-emphasized a bit and come with a bit of a warning. Nothing wrong with using them, but some awareness of the costs of these features will save a lot of work tracking down why a piece of code is so slow and will help a lot in preventing the impression that Scala is a purely academic language.

These "nice" features then shouldn't be the first thing a Scala novice reads about, even though they demonstrate the elegance of the language.

Bill

+1 rock-hard IDE support. When I developed in C# on .NET in a
corporate environment, I wanted to use F# (while it was beta) for
experimentation but the overhead of learning a new language alongside
IDE problems was too much. This also shows how hard it is to create
solid IDE integration since Microsoft controls Visual Studio and
releases both C# and F#.

I almost didn't use Scala because of how buggy ScalaIDE used to be.
It's a lot better now, but if it were as solid as Eclipse Java support
it'd be easy for someone to switch over and start adding Scala to
their projects.

Of course, rock-solid IDE support relates to the speed of the compiler!

Peace. Michael

On Sat, Nov 5, 2011 at 7:54 PM, William la Forge wrote:
> When you talk about real world, it is important that some of those very nice
> features of the language which are noticably slower than what a novice would
> expect should be de-emphasized a bit and come with a bit of a warning.
> Nothing wrong with using them, but some awareness of the costs of these
> features will save a lot of work tracking down why a piece of code is so
> slow and will help a lot in preventing the impression that Scala is a purely
> academic language.
>
> These "nice" features then shouldn't be the first thing a Scala novice reads
> about, even though they demonstrate the elegance of the language.
>
> Bill
>

My humble take on things...

1. The very real fear of Scala needs to be addressed. I cannot remember when I have ever seen such a controversial programming language and there just seems to be so many "Scala is too complex" concerns raised by too many people.
   
   
2. IDE Support. I still struggle with installing new versions of Scala with Eclipse and run into obscure errors that I would never tolerate in Java. To coin a phrase from Steve Jobs, when he was creating NeXT, "it just works." Scala in the IDE has to be close to as dependable as Java in the IDE - it just has to work - no exceptions, no excuses. Troubleshooting Scala plug-in problems is a very serious turn-off and makes the technology seem like it is still an academic curiosity.
   
   
3. Every time I delve into the Scaladocs I cringe at the complexity - it makes me feel stupid, as in, I am not intelligent enough for this.
   

Cheers, Eric

On 2011-11-05 2:51 PM, Ken McDonald wrote:

Even though this perhaps should be in scala-debate, I'm posting it here because:     1) It's really very important.     2) I've never seen a similar question posted.
This comes after reading a few of the "White elefant" posts.
The question is simple: List, in descending order of priority, what you feel needs to be addressed/fixed/whatever for Scala to succeed in the corporate (and hence IMHO ultimately the "real") world.
My list:
1. IDE Support. Scala must have rock-hard support in at least one IDE, and must have "pretty good" support in Eclipse, since it's the de facto standard. I don't think we're close yet. This alone could lose Scala the language wars.
2. Documentation. Scaladoc documentation needs to be expanded on greatly; at the same time Scaladoc itself could stand some enhancements, though exactly where is less obvious. I am complicit in this; I keep on intending to devote some serious time to enhancing scaladoc, and keep failing to do so, dangit!
3. A "Scala Cookbook". I'm amazed one isn't out or on the web already. I know, there's lots on stackoverflow and other sites but it's not the same. Scala is such a neat language, that a newcomer could easily get caught up and read for hours on a good cookbook, saying things like, "Wow, I can do it in just one line...".

And your suggestions? Recommended three or less because probably no one will read past the third one :-).


Ken

I think we really need SIQ or whatever we call it shipping as soon as possible.

Being able to say

here are the hundreds of lines of code you need to work with a database:

• you need some classes for ORM,
  
• dozens of annotations to describe the connection between them
• dozens of lines for simple type safe queries
• you need to know the quirks of Hibernate/JPA/EclipseLink and
  
• you probably still need to write SQL for more complex tasks

and here is how you do that in Scala, with a single line of code:

<example>

I think if we would have this, we have more or less won the whole enterprise stuff.

On Saturday, November 5, 2011 8:00:53 PM UTC-4, Andreas Flierl wrote:

(Also, I find "language wars" a strange term.)

A corporate environment with enough legacy (basically meaning "it's been around for a while") will almost certainly have a contingent of "defenders of the status quo" for any given technology that's had some success.  This is especially true if some influential individual(s) have linked their personal success to the success of this technology.  In a large enough environment there will likely be several of these groups, and they will fight with each other as much as with anything that's new.  On one hand this is all well and good, because change costs serious time and money, and often multiple overlapping technologies exist for a reason, so to the extent that fights stay on technical grounds they can be productive.  On the other hand, the fights often leave the rational world and exist almost entirely in the realm of politics.
A corporate environment with enough R&D investment in enabling technologies will almost certainly have one or more contingent of "evangelists of the new doodad."  Their aggressiveness will be roughly equivalent to the square of the amount of money invested in their doodad divided by the amount of success it has obtained.  On one hand this is can be productive, so long as the evangelists stick to technical grounds.  On the other hand, once a given technology has passed some degree of investment, the evangelism starts leaving the technical realm and becomes mostly political.
A "technology" could be a programming language, web framework, operating system, process, tool, semiconductor substrate, etc...
Now, the poor sap who has to actually create a product (meaning most engineers) has to contend with all these people who have opinions about what technologies he's supposed to use.  Furthermore, if he decides to toss his own competing technology into the mix because he thinks it fits his problem better than the ones associated with existing factions, he's going to face challenges from all sides.
If you're at a startup you probably don't have to worry about this much.  If you're at a Fortune 100 company (or one is your customer), you probably have more legacy hanging around than is humanly comprehendible and a fair bit of R&D, too.
If this poor sap is trying to use Scala, then he is "fighting a battle" for Scala in the "language wars."  Calling a bunch of nerds and managers arguing a "battle" in a "war" seems to involve a high amount of hyperbole, but it's a popular analogy none-the-less.
Anyway, if you're trying to use Scala, you could follow Tony's advice:http://blog.tmorris.net/i-cannot-use-language-x/#comments

Except that I think the barriers identified in this thread represent rational reasons for not using Scala at this time, so rational objections will inevitably be mixed in with all the political nonsense.  You should also remember the Golden Rule: He who has the gold makes the rules.  Sticking to principles works pretty well so long as the funding isn't cut off, so you can dismiss the irrational objections of others so long as the person with the gold is doing the same.
Of course, if you dismiss all these objections and run with Scala, and then the project fails, you run a high risk that Scala will be blamed regardless of actual fault, and you will be blamed as the person who ignored the wisdom of others and charged ahead with an immature technology.

that already exists, it's called db4o :)

Am 06.11.2011 14:16, schrieb Simon Ochsenreither:

I think we really need SIQ or whatever we call it shipping as soon as possible.

Being able to say

here are the hundreds of lines of code you need to work with a database:

• you need some classes for ORM,
  
• dozens of annotations to describe the connection between them
• dozens of lines for simple type safe queries
• you need to know the quirks of Hibernate/JPA/EclipseLink and
  
• you probably still need to write SQL for more complex tasks

and here is how you do that in Scala, with a single line of code:

<example>

I think if we would have this, we have more or less won the whole enterprise stuff.

Which no one uses ...

I certainly don't want to choose a database based on the API.

Am 06.11.2011 19:00, schrieb Simon Ochsenreither:
> Which no one uses ...
>
> I certainly don't want to choose a database based on the API.
you can always fall back to just saving hashmaps using an
objectoutputstream.

there is no solution which is both simple and efficient. you can have
one or the other.

On Sat, Nov 5, 2011 at 10:51 PM, Ken McDonald <ykkenmcd@gmail.com> wrote:

Even though this perhaps should be in scala-debate, I'm posting it here because:    1) It's really very important.    2) I've never seen a similar question posted.
This comes after reading a few of the "White elefant" posts.
The question is simple: List, in descending order of priority, what you feel needs to be addressed/fixed/whatever for Scala to succeed in the corporate (and hence IMHO ultimately the "real") world.
My list:

Your list is pretty much spot on as far as I can tell. Here's what we (Typesafe & EPFL) are doing about it.

 

1. IDE Support. Scala must have rock-hard support in at least one IDE, and must have "pretty good" support in Eclipse, since it's the de facto standard. I don't think we're close yet. This alone could lose Scala the language wars.

Typesafe now has 3 FTEs working on the Scala Eclipse IDE, plus 2 part-timers (one of them myself). None of this is paid for by anyone. We are doing this because we think its important, and universities can't do it. It's grunt work. Have you tried a recent beta of Scala IDE 2.0? If you have serious problems we'd like to see a ticket!
Also, I realize that even with this amount of committed resources we are still moving too slowly. So if you have time & expertise to contribute, this would be hugely appreciated! 

2. Documentation. Scaladoc documentation needs to be expanded on greatly; at the same time Scaladoc itself could stand some enhancements, though exactly where is less obvious. I am complicit in this; I keep on intending to devote some serious time to enhancing scaladoc, and keep failing to do so, dangit!

Heather Miller and Josh Suereth have recently released http://docs.scala-lang.org/. Now there's no more excuse not to contribute!

3. A "Scala Cookbook". I'm amazed one isn't out or on the web already. I know, there's lots on stackoverflow and other sites but it's not the same. Scala is such a neat language, that a newcomer could easily get caught up and read for hours on a good cookbook, saying things like, "Wow, I can do it in just one line...".

I think Cay Horstmann's excellent "Scala for the Impatient" comes close. A preprint of the first half is available for free from the typesafe.com site.
Cheers
 -- Martin

Did you have a look at the things happening with SIQ/...?

1. One definition of the database url and the credentials
2. .fromDb
3. Use the API known from the collection stuff

That's how it should be done.

  Hi, Ken
  Scala Cookbook , indeed a missing book.  I should extend your list with :
  - Scala Swing book  - Scala RIA?        Not bad is Scala Vaadin but no real good information.  - Scala Android developer book.
   Luc.

On Sat, Nov 5, 2011 at 22:51, Ken McDonald <ykkenmcd@gmail.com> wrote:

Even though this perhaps should be in scala-debate, I'm posting it here because:    1) It's really very important.    2) I've never seen a similar question posted.
This comes after reading a few of the "White elefant" posts.
The question is simple: List, in descending order of priority, what you feel needs to be addressed/fixed/whatever for Scala to succeed in the corporate (and hence IMHO ultimately the "real") world.
My list:
1. IDE Support. Scala must have rock-hard support in at least one IDE, and must have "pretty good" support in Eclipse, since it's the de facto standard. I don't think we're close yet. This alone could lose Scala the language wars.
2. Documentation. Scaladoc documentation needs to be expanded on greatly; at the same time Scaladoc itself could stand some enhancements, though exactly where is less obvious. I am complicit in this; I keep on intending to devote some serious time to enhancing scaladoc, and keep failing to do so, dangit!
3. A "Scala Cookbook". I'm amazed one isn't out or on the web already. I know, there's lots on stackoverflow and other sites but it's not the same. Scala is such a neat language, that a newcomer could easily get caught up and read for hours on a good cookbook, saying things like, "Wow, I can do it in just one line...".

And your suggestions? Recommended three or less because probably no one will read past the third one :-).


Ken

Rock solid IDE and maven support.
 
My experience so far is that the maven plugins all work absolutely fine - BUT so much emphasis (here and the lift forum) is around sbt that it is very easy to pick up the misunderstanding that you need to switch build tool.  That would be an automatic fail in many enviroments as maven is so heavily entrenched. Configuring builds is boring crap no-one wants to do twice.
 
For this reason, I'm making a deliberate point of not using sbt as I climb the scala learning curve. Should I find a bug, I'll be raising it very promptly !
 
There's some points in a similar question on linkedin that may be of relevance: Is scala the future of java

IDE support is far and away #1 for me. This comes from two standpoints:

1. I like a good IDE, even though Scala has made me less IDE-dependent than I used to be since I use the repl and other iterative tools so much more than I used to.
2. Even more important, when I am trying to get other developers to adopt Scala the first thing they want is a good IDE plugin. Getting others to try out Scala requires a gentle introduction. The first thing they want to do is write Java-as-Scala in their IDE of choice, not dive into FP. From what I have seen, the Eclipse plugin has come a long way towards making this happen (good job, team).

As an aside, it seems like "better IDE support" is frequently translated as "better Eclipse support." However, it seems that many experienced Scala developers use some sort of "Glorified Code Editor"+build tool to avoid many of the Eclipse headaches and to take advantage of the great features of mvn/sbt/fsc/etc. (they all take some learning, but are more powerful the deeper you dive). Most of the time I use NetBeans as a glorified code editor with some combination of sbt and/or mvn running in a console. To this end, I would really like to see better support for the NB plugin.
Just my $0.02.
-Mark

On Mon, 2011-11-07 at 10:50 -0500, Bastian, Mark wrote:

> I use NetBeans as a glorified code editor with some combination of sbt
> and/or mvn running in a console. To this end, I would really like to
> see better support for the NB plugin.

I'm also a NetBeans user and would love to see the NetBeans plugin
continue to evolve. Yes, I know... why don't I contribute? Well, as for
many of us I just don't have the personal resources to do that right
now.

I can't wait until retirement. Then I'll be able to get some real
programming done. :)

Peter

And your suggestions? Recommended three or less because probably no one will read past the third one :-).


Ken

I strongly agree with IDE as #1. However, I am skeptical about the value of better Scala Docs. Perhaps I am a pessimist, but while I agree that better Scala Doc would be incredibly useful to me as a developer, I think it means very little to CTOs and people who are on the fence about learning Scala. 
I think organizing meet ups and sharing success stories in the development community will be key. At a recent Scala meet up in Los Angeles, I met a small group of developers who are interested in getting started with Scala, but noone with in depth knowledge and success stories. I think that can be disheartening to new developers. In this respect, I think I can only rely on companies like TypeSafe getting out and help organize more meetups in major development hubs. 

--
---
John L Cheng

about Scala RIA

maybe ScalaFX is an interesting technology to consider
http://code.google.com/p/scalafx

btw: here is ScalaFX demo that I have developed
http://code.google.com/p/scalafx/source/browse/demo/scalafx/JumpingFrogsPuzzle.scala?spec=svn280146c3b01da9dac672e9b3bfedaddb00fe7a78&r=280146c3b01da9dac672e9b3bfedaddb00fe7a78

Luc

On Mon, Nov 7, 2011 at 10:56 AM, Luc Evers <lucevers@gmail.com> wrote:

  Hi, Ken
  Scala Cookbook , indeed a missing book.  I should extend your list with :
  - Scala Swing book  - Scala RIA?        Not bad is Scala Vaadin but no real good information.  - Scala Android developer book.
   Luc.

On Sat, Nov 5, 2011 at 22:51, Ken McDonald <ykkenmcd@gmail.com> wrote:

Even though this perhaps should be in scala-debate, I'm posting it here because:    1) It's really very important.    2) I've never seen a similar question posted.
This comes after reading a few of the "White elefant" posts.
The question is simple: List, in descending order of priority, what you feel needs to be addressed/fixed/whatever for Scala to succeed in the corporate (and hence IMHO ultimately the "real") world.
My list:
1. IDE Support. Scala must have rock-hard support in at least one IDE, and must have "pretty good" support in Eclipse, since it's the de facto standard. I don't think we're close yet. This alone could lose Scala the language wars.
2. Documentation. Scaladoc documentation needs to be expanded on greatly; at the same time Scaladoc itself could stand some enhancements, though exactly where is less obvious. I am complicit in this; I keep on intending to devote some serious time to enhancing scaladoc, and keep failing to do so, dangit!
3. A "Scala Cookbook". I'm amazed one isn't out or on the web already. I know, there's lots on stackoverflow and other sites but it's not the same. Scala is such a neat language, that a newcomer could easily get caught up and read for hours on a good cookbook, saying things like, "Wow, I can do it in just one line...".

And your suggestions? Recommended three or less because probably no one will read past the third one :-).


Ken

--
   __~O
  -\ <,
(*)/ (*)

reality goes far beyond imagination

think Scala's weakness (with regards to its adoption in the "Enterprise") is that it never projected/marketed  itself as such... the focus has always been on esoteric topics like bloom filters, actors etc. Now if we were to divide Enterprises into two broader categories (1) Traditional Enterprises (which are against any disruption/innovation...like most major financial institutions) (b) New/Innovative Enterprises (like Facebook, Google etc) ... Scala's appeal is to the second types of Enterprises. For first type of Enterprises you need many examples like "Duke's shopping cart" ;-) ... thus showing that Scala (Like Java/JEE) can answer the needs for presentation and business tier and Web Services ... so yes ..such a book is required.

my two cents worth

cheers

On Mon, Nov 7, 2011 at 1:49 PM, Luc Duponcheel <luc.duponcheel@gmail.com> wrote:

about Scala RIA

maybe ScalaFX is an interesting technology to consider
http://code.google.com/p/scalafx

btw: here is ScalaFX demo that I have developed
http://code.google.com/p/scalafx/source/browse/demo/scalafx/JumpingFrogsPuzzle.scala?spec=svn280146c3b01da9dac672e9b3bfedaddb00fe7a78&r=280146c3b01da9dac672e9b3bfedaddb00fe7a78

Luc

On Mon, Nov 7, 2011 at 10:56 AM, Luc Evers <lucevers@gmail.com> wrote:

  Hi, Ken
  Scala Cookbook , indeed a missing book.  I should extend your list with :
  - Scala Swing book  - Scala RIA?        Not bad is Scala Vaadin but no real good information.  - Scala Android developer book.
   Luc.

On Sat, Nov 5, 2011 at 22:51, Ken McDonald <ykkenmcd@gmail.com> wrote:

Even though this perhaps should be in scala-debate, I'm posting it here because:    1) It's really very important.    2) I've never seen a similar question posted.
This comes after reading a few of the "White elefant" posts.
The question is simple: List, in descending order of priority, what you feel needs to be addressed/fixed/whatever for Scala to succeed in the corporate (and hence IMHO ultimately the "real") world.
My list:
1. IDE Support. Scala must have rock-hard support in at least one IDE, and must have "pretty good" support in Eclipse, since it's the de facto standard. I don't think we're close yet. This alone could lose Scala the language wars.
2. Documentation. Scaladoc documentation needs to be expanded on greatly; at the same time Scaladoc itself could stand some enhancements, though exactly where is less obvious. I am complicit in this; I keep on intending to devote some serious time to enhancing scaladoc, and keep failing to do so, dangit!
3. A "Scala Cookbook". I'm amazed one isn't out or on the web already. I know, there's lots on stackoverflow and other sites but it's not the same. Scala is such a neat language, that a newcomer could easily get caught up and read for hours on a good cookbook, saying things like, "Wow, I can do it in just one line...".

And your suggestions? Recommended three or less because probably no one will read past the third one :-).


Ken

--
   __~O
  -\ <,
(*)/ (*)

reality goes far beyond imagination

If only Scala were *not* being adopted in major financial
institutions, like it is, that argument would be stronger.

Alas, someone once divided financial institutions between loaning
banks and investment banks, where the first are very conservative and
the second will try anything that can give it an edge.

On Tue, Nov 8, 2011 at 13:09, Arif Mustafa wrote:
> think Scala's weakness (with regards to its adoption in the "Enterprise") is
> that it never projected/marketed  itself as such... the focus has always
> been on esoteric topics like bloom filters, actors etc. Now if we were to
> divide Enterprises into two broader categories (1) Traditional Enterprises
> (which are against any disruption/innovation...like most major financial
> institutions) (b) New/Innovative Enterprises (like Facebook, Google etc) ...
> Scala's appeal is to the second types of Enterprises. For first type of
> Enterprises you need many examples like "Duke's shopping cart" ;-) ... thus
> showing that Scala (Like Java/JEE) can answer the needs for presentation and
> business tier and Web Services ... so yes ..such a book is required.
>
> my two cents worth
>
> cheers
>
> On Mon, Nov 7, 2011 at 1:49 PM, Luc Duponcheel
> wrote:
>>
>> about Scala RIA
>>
>> maybe ScalaFX is an interesting technology to consider
>> http://code.google.com/p/scalafx
>>
>> btw: here is ScalaFX demo that I have developed
>>
>> http://code.google.com/p/scalafx/source/browse/demo/scalafx/JumpingFrogs...
>>
>> Luc
>>
>> On Mon, Nov 7, 2011 at 10:56 AM, Luc Evers wrote:
>>>
>>>   Hi, Ken
>>>   Scala Cookbook , indeed a missing book.
>>>   I should extend your list with :
>>>   - Scala Swing book
>>>   - Scala RIA?
>>>        Not bad is Scala Vaadin but no real good information.
>>>   - Scala Android developer book.
>>>    Luc.
>>>
>>> On Sat, Nov 5, 2011 at 22:51, Ken McDonald wrote:
>>>>
>>>> Even though this perhaps should be in scala-debate, I'm posting it here
>>>> because:
>>>>     1) It's really very important.
>>>>     2) I've never seen a similar question posted.
>>>> This comes after reading a few of the "White elefant" posts.
>>>> The question is simple: List, in descending order of priority, what you
>>>> feel needs to be addressed/fixed/whatever for Scala to succeed in the
>>>> corporate (and hence IMHO ultimately the "real") world.
>>>> My list:
>>>> 1. IDE Support. Scala must have rock-hard support in at least one IDE,
>>>> and must have "pretty good" support in Eclipse, since it's the de facto
>>>> standard. I don't think we're close yet. This alone could lose Scala the
>>>> language wars.
>>>> 2. Documentation. Scaladoc documentation needs to be expanded on
>>>> greatly; at the same time Scaladoc itself could stand some enhancements,
>>>> though exactly where is less obvious. I am complicit in this; I keep on
>>>> intending to devote some serious time to enhancing scaladoc, and keep
>>>> failing to do so, dangit!
>>>> 3. A "Scala Cookbook". I'm amazed one isn't out or on the web already. I
>>>> know, there's lots on stackoverflow and other sites but it's not the same.
>>>> Scala is such a neat language, that a newcomer could easily get caught up
>>>> and read for hours on a good cookbook, saying things like, "Wow, I can do it
>>>> in just one line...".
>>>>
>>>> And your suggestions? Recommended three or less because probably no one
>>>> will read past the third one :-).
>>>>
>>>>
>>>> Ken
>>
>>
>>
>> --
>>    __~O
>>   -\ <,
>> (*)/ (*)
>>
>> reality goes far beyond imagination
>>
>
>

On Tuesday, 8 November 2011 19:52:55 UTC, Daniel Sobral wrote:

If only Scala were *not* being adopted in major financial
institutions, like it is, that argument would be stronger.

Alas, someone once divided financial institutions between loaning
banks and investment banks, where the first are very conservative and
the second will try anything that can give it an edge.

Spot on. From what I can see, Scala is rapidly gaining traction in most major financial orgs. The devs love it and don't mind that the IDE's aren't great right now, the benefits of the language outweigh the short-term problem of good IDEs.
What needs to happen is for everyone to stop worrying about a problem that doesn't really exist.

For my part, I'm seeing a lot of adoption in media.  Including broadcast, papers, online gaming, social, etc, etc.

Basically, lots of people who're after an edge in being more responsive to the market than their competition, so not just investment banking :)

Banks are just more visible because they tend to pay more for support contacts, and use recruiters who are far spammier on LinkedIn and the like.

On Nov 8, 2011 9:09 PM, "Channing Walton" <channingwalton@mac.com> wrote:

On Tuesday, 8 November 2011 19:52:55 UTC, Daniel Sobral wrote:

If only Scala were *not* being adopted in major financial
institutions, like it is, that argument would be stronger.

Alas, someone once divided financial institutions between loaning
banks and investment banks, where the first are very conservative and
the second will try anything that can give it an edge.

Spot on. From what I can see, Scala is rapidly gaining traction in most major financial orgs. The devs love it and don't mind that the IDE's aren't great right now, the benefits of the language outweigh the short-term problem of good IDEs.
What needs to happen is for everyone to stop worrying about a problem that doesn't really exist.

@Kevin spot on... "broadcast, papers, online gaming, social, etc, etc." ...and thats where I have observed the adoption rate of Scala to be high ... and ..."Banks"  (Tradional model, who would rather pay IBM'ers and/or Tata Consultancy etc to run their IT) have low adoption rate of Scala is low.

On Tue, Nov 8, 2011 at 4:28 PM, Kevin Wright <kev.lee.wright@gmail.com> wrote:

For my part, I'm seeing a lot of adoption in media.  Including broadcast, papers, online gaming, social, etc, etc.

Basically, lots of people who're after an edge in being more responsive to the market than their competition, so not just investment banking :)

Banks are just more visible because they tend to pay more for support contacts, and use recruiters who are far spammier on LinkedIn and the like.

On Nov 8, 2011 9:09 PM, "Channing Walton" <channingwalton@mac.com> wrote:

On Tuesday, 8 November 2011 19:52:55 UTC, Daniel Sobral wrote:

If only Scala were *not* being adopted in major financial
institutions, like it is, that argument would be stronger.

Alas, someone once divided financial institutions between loaning
banks and investment banks, where the first are very conservative and
the second will try anything that can give it an edge.

Spot on. From what I can see, Scala is rapidly gaining traction in most major financial orgs. The devs love it and don't mind that the IDE's aren't great right now, the benefits of the language outweigh the short-term problem of good IDEs.
What needs to happen is for everyone to stop worrying about a problem that doesn't really exist.

> The question is simple: List, in descending order of priority, what you
> feel needs to be addressed/fixed/whatever for Scala to succeed in the
> corporate (and hence IMHO ultimately the "real") world.

I'll only list 1: reliability.

As a software development manager, I recently shied away from Scala in
a newly starting project mainly because I judged the compiler to be
unreliable.

As for IDE support, while it is very important (even essential to
attract the Java lot), it is only an absolute must because the correct/
compile/run cycle is so slow. PHP, for example, is quite successful in
spite of having extremely poor IDE/tooling support.

--
Salut,

Jordi.

On Tue, Nov 8, 2011 at 9:48 PM, Jordi Salvat i Alabart <jordi.salvat.i.alabart@gmail.com> wrote:

mainly because I judged the compiler to be
unreliable.

Why?

IDE Support would be the most important issue for me. I use Netbeans to write the code and SBT on a terminal to compile and run. I really do not want to use Eclipse, I don't like it as a whole (I don't have a problem with the Scala IDE itself). How hard would it be to have the Eclipse plugin work in Netbeans? It supports OSGI so maybe it could re-use a lot of stuff. Off course integration with the editor itself is definitely going to be different but Caoyuan has gone a long way with the Scala plugin for NB and there should be enough stuff already figured out there.

I started using Scala in the "real world" more than 3 years ago, I think 4. We had a custom server over TCP, hundreds of clients, several databases and many other legacy servers using different custom communication protocols as well as some desktop applications for internal use and for clients... all of that was done in Java. One day we discovered Scala, thought it was cool, bought the first book the day it came out and started using it right away when we noticed we could mix Scala code with our current code base without much trouble. We ended up converting most of the code to Scala and only produced new code with it. We never looked back. I don't work there anymore but they still use Scala. There is no compelling reason not to use it in the "real world".

I never had a problem with documentation or the available books regarding the language itself. I never minded googling for stuff or browsing stack overflow. I never had a problem asking questions on the mailing list either.

On 11/09/2011 11:31 AM, Tomás Lázaro wrote:

CALVYeJi09fHpdndfJCcLQGUvD0ZzbvSiwZ7bhDdGB3VcRggKBw [at] mail [dot] gmail [dot] com" type="cite">IDE Support would be the most important issue for me. I use Netbeans to write the code and SBT on a terminal to compile and run. I really do not want to use Eclipse, I don't like it as a whole (I don't have a problem with the Scala IDE itself). How hard would it be to have the Eclipse plugin work in Netbeans? It supports OSGI so maybe it could re-use a lot of stuff. Off course integration with the editor itself is definitely going to be different but Caoyuan has gone a long way with the Scala plugin for NB and there should be enough stuff already figured out there.

I started using Scala in the "real world" more than 3 years ago, I think 4. We had a custom server over TCP, hundreds of clients, several databases and many other legacy servers using different custom communication protocols as well as some desktop applications for internal use and for clients... all of that was done in Java. One day we discovered Scala, thought it was cool, bought the first book the day it came out and started using it right away when we noticed we could mix Scala code with our current code base without much trouble. We ended up converting most of the code to Scala and only produced new code with it. We never looked back. I don't work there anymore but they still use Scala. There is no compelling reason not to use it in the "real world".

I never had a problem with documentation or the available books regarding the language itself. I never minded googling for stuff or browsing stack overflow. I never had a problem asking questions on the mailing list either.

The problem with eclipse scala-ide I found
1. Performance, too much heap memory usage , too slow.
2. Syntax color, I always see only two color, that is somewhat boring.
3.Coding express is not good enough. Code completion sometimes not available.

The problem with Netbeans
1.Swing seems not work well with linux.

2011/11/9 Tomás Lázaro <tlazaro18@gmail.com>

IDE Support would be the most important issue for me. I use Netbeans to write the code and SBT on a terminal to compile and run. I really do not want to use Eclipse, I don't like it as a whole (I don't have a problem with the Scala IDE itself). How hard would it be to have the Eclipse plugin work in Netbeans? It supports OSGI so maybe it could re-use a lot of stuff. Off course integration with the editor itself is definitely going to be different but Caoyuan has gone a long way with the Scala plugin for NB and there should be enough stuff already figured out there.

I think Caoyuan is already using the presentation compiler, which is at the core of the Eclipse IDE. And so is ENSIME. Hopefully it will only take minor upgrade works to the latest version to get many improvements. The other large trend is that we are going to put more functionality out of the Eclipse SDT and into the presentation compiler. That should also help the other plugins. 
Cheers
 -- Martin

Thomas, Wow.

Have you been using Netbeans to compile real world projects? My
experience of Scala Netbeans integration is:
The guy who was maintaining the plug-in left 2 years ago and that it
currently chokes on anything more serious than Hello World.

Scala-IDE, the Eclipse plugin while not quite there compared to
Eclipse's Java support is still WAAAEEEEY ahead of the competition.

And yes, I use IDEA for ActionScript but it's Scala support is at
best, enthusiast grade.

I say this having written an approx 30,000 LOC Scala project last year
all of which parses and compiles (slowly) using Scala-IDE and/or
Maven.

Regards,

Bryan Hunt

On 2011, Nov 9, at 11:51 AM, bryan hunt wrote:
> Scala-IDE, the Eclipse plugin while not quite there compared to
> Eclipse's Java support is still WAAAEEEEY ahead of the competition.
>
> And yes, I use IDEA for ActionScript but it's Scala support is at
> best, enthusiast grade.

Oh, I dont know. My team all use intellij since last year or so. Eclipse just did not work out for us - it was too buggy.

We all seem to think intellij is good enough. So do the consultants I know around here who actively use Scala.

YMMV.

Geir

You may be having problem with inference. Reducing inference by annotating types can help reduce ram consumption and compilation time significantly. This is mostly true for multi-level inference that has to go through several layers of parentheses.
For example: val x = A(B(C(a, b, c)))If the type of x is something like A[X[Y[Z]]], then you are better off declaring it: val x = A[X[Y[Z]]](B(C(a, b, c)))
It may take a bit of extra work, at least with large existing projects, but it will make everything compile in reasonable amount of time.
This problem has been fixed for command line compilations, but it still makes Eclipse IDE eat up all ram and slow down to a crawl. My guess it will be fixed in Eclipse sometime soon as well.

On Wed, Nov 9, 2011 at 4:51 AM, bryan hunt <sentimental.bryan@gmail.com> wrote:

Thomas, Wow.

Have you been using Netbeans to compile real world projects? My
experience of Scala Netbeans integration is:
The guy who was maintaining the plug-in left 2 years ago and that it
currently chokes on anything more serious than Hello World.

Scala-IDE, the Eclipse plugin while not quite there compared to
Eclipse's Java support is still WAAAEEEEY ahead of the competition.

And yes, I use IDEA for ActionScript but it's Scala support is at
best, enthusiast grade.

I say this having written an approx 30,000 LOC Scala project last year
all of which parses and compiles (slowly) using Scala-IDE and/or
Maven.

Regards,

Bryan Hunt

On Sun, Nov 6, 2011 at 2:08 AM, Erik Engbrecht wrote:
> I think you need to define what "rock-hard support" in an IDE means.  For
> example, the Eclipse plugin keeps on moving forward in terms of
> functionality, but where exactly is the version of it for Scala 2.9.1?

If you refer to the download page, the version number is indeed 2.9.2,
because of changes to the presentation compiler. Since the
presentation compiler is part of the compiler, we need an updated
version to take advantage of the new features. However, since
yesterday we ship with the standard 2.9.1 library, so you should not
feel any difference.

The Eclipse IDE is really low on features. We're basically focusing on
completion, hyperlinking and interactive error reporting. IntelliJ has
been a lot more active in adding new features, though, so maybe you
mixed them.

>  It
> also consumes a ton of memory and tends to "just sit there" a lot.  It's
> basically unusable in my work environment, while the version quite a while
> back, while buggier and with far fewer features, was usable.

What versions are we talking about, or when did you try Eclipse last?
This sounds really surprising, since my own experience, and most
people that cared to try Eclipse one year ago, agreed that it's much
more usable now.

>I think the
> NetBeans plugin also took a step sideways when it integrated with the
> presentation compiler.  In my opinion the only really usable IDE for Scala
> right now is Emacs+ENSIME.  Maybe the ENSIME integrations for Vim and JEdit

ENSIME uses the presentation compiler, so it can't be much better
(assuming the presentation compiler is the problem).

iulian

> work well, too.  I'm fine with Emacs, but I wouldn't try to force in on
> anyone.
>
>
>

On Wed, 2011-11-09 at 05:51 -0500, bryan hunt wrote:

> Thomas, Wow.
>
> Have you been using Netbeans to compile real world projects? My
> experience of Scala Netbeans integration is:
> The guy who was maintaining the plug-in left 2 years ago and that it
> currently chokes on anything more serious than Hello World.

I'm using NetBeans on a Scala project. It's not a huge project, but
definitely larger than "Hello, World." The Scala plugin is working fine.
It's a bit quirky but it gets the job done. I should say that I'm using
the latest version from the git repository compiled from source. The
person maintaining the plugin is still active with it, although activity
is light. The last commit was maybe a couple of months ago (I haven't
tried a pull recently though).

Peter

> I'm using NetBeans on a Scala project. It's not a huge project, but
> definitely larger than "Hello, World." The Scala plugin is working fine.
> It's a bit quirky but it gets the job done. I should say that I'm using
> the latest version from the git repository compiled from source.

For the curious, the git repo is https://github.com/dcaoyuan/nbscala

I find this very hard to believe. The JetBrains folks have been
working steadily on their plugin and they have made tremendous
progress in features and usability. In the beginning, it's syntax
parsing was often a bit sketchy but those problems have pretty much
disappeared and all of the IDEA niceties have started showing up.

I have found that IDEA vs Eclipse is a chocolate/vanilla or
purple/green question. I've had to use Eclipse a couple of times and I
just didn't like it.

Donald

On Wed, Nov 9, 2011 at 5:51 AM, bryan hunt wrote:
>
> Scala-IDE, the Eclipse plugin while not quite there compared to
> Eclipse's Java support is still WAAAEEEEY ahead of the competition.
>
> And yes, I use IDEA for ActionScript but it's Scala support is at
> best, enthusiast grade.

On 11/09/2011 01:04 PM, Geir Hedemark wrote:

0DB7631D-8B6D-423D-89DB-8949CF92AAF2 [at] gmail [dot] com" type="cite">

On 2011, Nov 9, at 11:51 AM, bryan hunt wrote:

Scala-IDE, the Eclipse plugin while not quite there compared to
Eclipse's Java support is still WAAAEEEEY ahead of the competition.

And yes, I use IDEA for ActionScript but it's Scala support is at
best, enthusiast grade.

Oh, I dont know. My team all use intellij since last year or so. Eclipse just did not work out for us - it was too buggy.

We all seem to think intellij is good enough. So do the consultants I know around here who actively use Scala.

YMMV.

Agreed.
It's not just how the IDE understands Scala itself, it's also about how the IDE integrates other features of the IDE in Scala-files/projects. IDEA is just fantastic here. My JPA-model written in Scala shows up in the "Show ER-diagram", my JPQL is highlighted correctly to give me code-completion in my JPQL working with JPA-entities written in Scala. SQL code-completion in scala-classes. Spring-integration for Scala is *almost* just as good as in JAVA. The list is long...

I use IDEA-11 EAP with the latest scala-plugin and I really don't consider IDE-support at all to be an issue anymore.

I only have one thing which I think is important to succeed in the corporate world:
1. Binary compatibility among version-upgrades

I have another nit-picking-list though:-)
- JAVA interoperability (yes, there are cases where they don't play nicely [1][2])
- Consistent handling of "val" (finals) [3]

[1] Should be able to use JAVA-enums and other "public static final" in annotation-parameters.
[2] https://issues.scala-lang.org/browse/SI-4549
[3] http://stackoverflow.com/questions/7762838/scala-forward-references-why-does-this-code-compile

-- 
Andreas Joseph Krogh  - mob: +47 909 56 963
Senior Software Developer / CTO - OfficeNet AS - http://www.officenet.no
Public key: http://home.officenet.no/~andreak/public_key.asc

Le 05/11/2011 22:51, Ken McDonald a écrit :
[...]
> The question is simple: List, in descending order of priority, what you
> feel needs to be addressed/fixed/whatever for Scala to succeed in the
> corporate (and hence IMHO ultimately the "real") world.
>
> My list:
>
> 1. IDE Support. [...]
>
> 2. Documentation. [...]
>
> 3. A "Scala Cookbook". [...]
>

My shop list would be:

1. A compiler as fast as the java one. And a modular scala-lib. Yeah, I
now...
2. An as powerful and great library to deals with immutable data
structure modification as the collection one is for collection (lens,
zipper, etc)
3. Being able to use Scala for scripting. Than implies 1 and a better
starting time + something to include automagically dependencies,
essentially. And it will be so good to show how cool and powerful Scala
is in some little lines. I really would *love* to convert the sysadmin I
work with to use it as possible replacement of perl or python (or bash),
but for now, the lauching time are killing that possibility.

IDE support is reaching the point where Eclipse does not go in my way
anymore and is of real help for boring things (imports, completion of
long method names, formatting, things like that).

Documentation is OK, especially with so many books and blogs post online.

A "Scala Cookbook" could be intersting, but I would much more prefer a
repository of Scala scripts, if they were usable for real.

Cheers,

On Wed, Nov 9, 2011 at 21:30, Francois Armand wrote:
> Le 05/11/2011 22:51, Ken McDonald a écrit :
> [...]
>>
>> The question is simple: List, in descending order of priority, what you
>> feel needs to be addressed/fixed/whatever for Scala to succeed in the
>> corporate (and hence IMHO ultimately the "real") world.
>>
>> My list:
>>
>> 1. IDE Support. [...]
>>
>> 2. Documentation. [...]
>>
>> 3. A "Scala Cookbook". [...]
>>
>
>
> My shop list would be:
>
> 1. A compiler as fast as the java one. And a modular scala-lib. Yeah, I
> now...
> 2. An as powerful and great library to deals with immutable data structure
> modification as the collection one is for collection (lens, zipper, etc)
> 3. Being able to use Scala for scripting. Than implies 1 and a better
> starting time + something to include automagically dependencies,
> essentially. And it will be so good to show how cool and powerful Scala is
> in some little lines. I really would *love* to convert the sysadmin I work
> with to use it as possible replacement of perl or python (or bash), but for
> now, the lauching time are killing that possibility.

You mean like this?

#!/usr/bin/env scalas
!#

/***
scalaVersion := "2.9.0-1"

libraryDependencies ++= Seq(
"net.databinder" %% "dispatch-twitter" % "0.8.3",
"net.databinder" %% "dispatch-http" % "0.8.3"
)
*/

import dispatch.{ json, Http, Request }
import dispatch.twitter.Search
import json.{ Js, JsObject }

def process(param: JsObject) = {
val Search.text(txt) = param
val Search.from_user(usr) = param
val Search.created_at(time) = param

"(" + time + ")" + usr + ": " + txt
}

Http.x((Search("#scala") lang "en") ~> (_ map process foreach println))

See https://github.com/harrah/xsbt/wiki/Scripts. Works like a charm.

>
> IDE support is reaching the point where Eclipse does not go in my way
> anymore and is of real help for boring things (imports, completion of long
> method names, formatting, things like that).
>
> Documentation is OK, especially with so many books and blogs post online.
>
> A "Scala Cookbook" could be intersting, but I would much more prefer a
> repository of Scala scripts, if they were usable for real.
>
> Cheers,
>
> --
> Francois Armand
> http://fanf42.blogspot.com
>

Le 10/11/2011 00:39, Daniel Sobral a écrit :
> On Wed, Nov 9, 2011 at 21:30, Francois Armand wrote:
[...]
>> 3. Being able to use Scala for scripting. Than implies 1 and a better
>> starting time + something to include automagically dependencies,
>> essentially. And it will be so good to show how cool and powerful Scala is
>> in some little lines. I really would *love* to convert the sysadmin I work
>> with to use it as possible replacement of perl or python (or bash), but for
>> now, the lauching time are killing that possibility.
>
> You mean like this?
>
[...]
> See https://github.com/harrah/xsbt/wiki/Scripts. Works like a charm.
>

Exactly like this, thanks for pointing it !

Now, it only remains compiler speed and starting time :)

(wow, that's really cool, what a pity I missed that... Does it exists
for a long time ?)

I found one bug in the compiler and 1 in the Language Spec in a few hours of tinkering around (well, actually I found more, but I was playing with experimental features so I didn't count those).

As a result of that research I fell in love with the language and have been playing with it in personal projects since. A couple of months later, I've filed a total of 21 bug reports (my username is "jsalvata" in Jira, if you want to see the listing).

Of course some of those may be false positives, and many are at the edges (playing with Xexperimental, problems internal to the compiler, ...), but I still stand by my initial assessment: Scala is not ready for the industry at large, except possibly for specially complex problems where it provides significant advantage (such as Scala's flagship use in industry, Twitter's engine)

--
Salut,

Jordi.

2011/11/9 Naftoli Gugenheim <naftoligug@gmail.com>

On Tue, Nov 8, 2011 at 9:48 PM, Jordi Salvat i Alabart <jordi.salvat.i.alabart@gmail.com> wrote:

mainly because I judged the compiler to be
unreliable.

Why?

On 9 Nov 2011, at 20:21, Andreas Joseph Krogh wrote:

> I use IDEA-11 EAP with the latest scala-plugin and I really don't consider IDE-support at all to be an issue anymore.

This is also my experience too although I use IDEA-11 and Eclipse. The only thing that irritates me in IDEA is that one still needs to do a full compile occasionally to see if the red lines in editors are real problems or not. But a couple of bugs I submitted recently were fixed in 1 day which made the experience much better.

The IDE is simply not an issue any more, certainly not one that should prevent anyone from benefitting from Scala.

Its all good!
On 8 Nov 2011, at 21:28, Kevin Wright wrote:

For my part, I'm seeing a lot of adoption in media.  Including broadcast, papers, online gaming, social, etc, etc.

Basically, lots of people who're after an edge in being more responsive to the market than their competition, so not just investment banking :)

Banks are just more visible because they tend to pay more for support contacts, and use recruiters who are far spammier on LinkedIn and the like.

On Nov 8, 2011 9:09 PM, "Channing Walton" <channingwalton@mac.com> wrote:

On Tuesday, 8 November 2011 19:52:55 UTC, Daniel Sobral wrote:

If only Scala were *not* being adopted in major financial
institutions, like it is, that argument would be stronger.

Alas, someone once divided financial institutions between loaning
banks and investment banks, where the first are very conservative and
the second will try anything that can give it an edge.

Spot on. From what I can see, Scala is rapidly gaining traction in most major financial orgs. The devs love it and don't mind that the IDE's aren't great right now, the benefits of the language outweigh the short-term problem of good IDEs.
What needs to happen is for everyone to stop worrying about a problem that doesn't really exist.

> On Sun, Nov 6, 2011 at 2:08 AM, Erik Engbrecht <erik.engbrecht@gmail.com> wrote:
> > I think you need to define what "rock-hard support" in an IDE means. 

Here are some of the things that IDEA can do for me in Java files that have not-much to do with refactoring/navigation. These things will (I suspect) come as a *big* surprise to the people who are used to VIM and emacs but they are the sorts of things that save a lot of time.
1. Intelligent string parsing. IDEA can tell me when my format string passed to String.format is malformed wrt to the types of the arguments I pass to it (it is highlighted as a bug). It can render *strings* which represent SQL or regex such that they have coloring appropriate to SQL (different color keywords etc).
2. Auto-complete *within Strings*. In a String value in my program, IDEA detects whether the string looks like a path and can auto-complete within the string. For example, if I am creating an ImageIcon with the path to an image resource, IDEA can offer me the choices of images within the directory. It even pops up an *image viewer* as I scroll down the options, even though all I have written is something like: "/path/to/images/ CTRL-SPACE"
3. Color choosing. IDEA spots when a java.awt.Color is created in my program and offers a little square of that color in the sidebar. Clicking on the square gives me a color picker which I can then use to change the color's parameters (requires enterprise edition)
4. Warnings: IDEA warns me when a piece of code is redundant, or when a return value or parameter is not used. It tells me when a method call is being placed against something that might reasonably be null (under certain circumstances), and when a code block (like an if-statement) can be simplified, or its sense reversed. etc etc.
5. When I perform a refactoring, IDEA spots whether other snippets of code in the same file are duplicates (and hence can also be replaced by the refactored item). Refactoring options are many; allowing me to extract fields, constants, parameters, perform renamings which are detected across other file types (such as properties files or XML config files).
6. I can navigate to my types and auto-complete methods and class names from within config or properties files. I can, from within a config file, get a popup of the constructor-params which a type takes. etc etc.
Admittedly these individually are minor things (and a few are already possible with the scala IDEA plugin) but they do add up in such a way that stepping into scala-land can feel less enterprisey (not always a bad thing). Obviously I feel that, on balance, scala is still way ahead in the productivity stakes - but you can at least imagine the reaction of developers when *features that they are used to and use every day* are no longer available.
Chris

-------- Original-Nachricht --------
> Datum: Thu, 10 Nov 2011 10:25:31 +0000
> Von: Chris Marshall
> An: scala-user@googlegroups.com
> Betreff: RE: [scala-user] Re: What is highest priority for Scala to succeed in corporate world (Should be in scala-debate?) ?

>
> > On Sun, Nov 6, 2011 at 2:08 AM, Erik Engbrecht
> wrote:
> > > I think you need to define what "rock-hard support" in an IDE means.
>
> Here are some of the things that IDEA can do for me in Java files that
> have not-much to do with refactoring/navigation. These things will (I suspect)
> come as a *big* surprise to the people who are used to VIM and emacs but
> they are the sorts of things that save a lot of time.
> 1. Intelligent string parsing. IDEA can tell me when my format string
> passed to String.format is malformed wrt to the types of the arguments I pass
> to it (it is highlighted as a bug). It can render *strings* which represent
> SQL or regex such that they have coloring appropriate to SQL (different
> color keywords etc).

it works for nested languages in general (sql, hibernate query stuff). for example, i have native javascript inside my gwt java classes. idea offers full support for these snippets. i pity my eclipse co-workers every time :)

> 2. Auto-complete *within Strings*. In a String value in my program, IDEA
> detects whether the string looks like a path and can auto-complete within
> the string. For example, if I am creating an ImageIcon with the path to an
> image resource, IDEA can offer me the choices of images within the directory.

really helpful inside xml files

> It even pops up an *image viewer* as I scroll down the options, even
> though all I have written is something like: "/path/to/images/ CTRL-SPACE"
> 3. Color choosing. IDEA spots when a java.awt.Color is created in my
> program and offers a little square of that color in the sidebar. Clicking on the
> square gives me a color picker which I can then use to change the color's
> parameters (requires enterprise edition)

it also shows the color in the debugger.

> 4. Warnings: IDEA warns me when a piece of code is redundant, or when a
> return value or parameter is not used. It tells me when a method call is
> being placed against something that might reasonably be null (under certain
> circumstances), and when a code block (like an if-statement) can be
> simplified, or its sense reversed. etc etc.
> 5. When I perform a refactoring, IDEA spots whether other snippets of code
> in the same file are duplicates (and hence can also be replaced by the
> refactored item). Refactoring options are many; allowing me to extract fields,
> constants, parameters, perform renamings which are detected across other
> file types (such as properties files or XML config files).
> 6. I can navigate to my types and auto-complete methods and class names
> from within config or properties files. I can, from within a config file, get
> a popup of the constructor-params which a type takes. etc etc.
> Admittedly these individually are minor things (and a few are already
> possible with the scala IDEA plugin) but they do add up in such a way that
> stepping into scala-land can feel less enterprisey (not always a bad thing).
> Obviously I feel that, on balance, scala is still way ahead in the
> productivity stakes - but you can at least imagine the reaction of developers when
> *features that they are used to and use every day* are no longer available.
> Chris

Ok, an example, one of many, Scala-Query (the project) is not parsed
correctly by Intellij.

Having spent a number of years on theserverside.com, I'm well versed
in the world of AstroTurfing(TM).

I am for the record, a fully paid up, subscribed Intellij Enterprise
customer, but I keep a copy of Eclipse for when I need to do Scala.

Intellij cannot accurately parse more than about 20 (Scala) classes
without crapping itself, Java and Flex support are outstanding, even
if the UI is a bit sluggish.

Regards,

Bryan Hunt

On Nov 9, 12:04 pm, Geir Hedemark wrote:
> On 2011, Nov 9, at 11:51 AM, bryan hunt wrote:
>
> > Scala-IDE, the Eclipse plugin while not quite there compared to
> > Eclipse's Java support is still WAAAEEEEY ahead of the competition.
>
> > And yes, I use IDEA for ActionScript but it's Scala support is at
> > best, enthusiast grade.
>
> Oh, I dont know. My team all use intellij since last year or so. Eclipse just did not work out for us - it was too buggy.
>
> We all seem to think intellij is good enough. So do the consultants I know around here who actively use Scala.
>
> YMMV.
>
> Geir