Plugin idea for web development

Not many people here probably do a lot of straight HTML programming, so maybe
this won't interest many of you. However, I think IDEA is in a position to
become a platform for Dreamweaver-like editing.

I last used Dreamweaver version MX 2004, and it sucked pretty bad. You could
edit the source and the graphical view (powered by Opera) and they would sync
up, and the selection would sync up and things like that, but you had to click a
resync button when you were done editing the source, and it didn't update live
in most cases. I think this is because Macromedia doesn't have the kinds of
lexing and parsing code in their editor that modern Java IDE's have.

I think an awesome plugin would be to take advantage of the fact that IDEA
always has a HTML page's DOM in memory, and it's completely up to date with the
editor. You could use JDIC browser component to embed mozilla browser component
in a toolwindow and keep it updated on every keypress. From there I think it
could expand to support editing in the mozilla component and keeping changes to
the dom in sync with IDEA's editor.

65 comments
Comment actions Permalink

I think it has to be about Ruby as well. Otherwise, you can simply wipe out the advantage of Rails by using Trails in Java. I don't think most Ruby fanatics would be ok with that. :)

0
Comment actions Permalink

Language performance is a non-issue for 99% of webapps. Most of the time

is spent marshalling data and DB access anyway. Plus, Ruby is still
interpreted. When Ruby 2.0 comes out with a compiler, Ruby performance
will drastically improve. (Not where the puck is...) Rails is not
targeting (currently) heavily clustered environments, it's targeting the
other 99% of webapps (the PHPs, Pythons, JSPs, ASP.NETs). And in any
case, scalability is currently good enough and will only get better.

You are vastly oversimplifying things. Language performance IS an issue when it's simply too slow. The fact that they are working on a compiler is meaningless because 1) it's not available now, and 2) we just don't know how fast it will be. Finally, some webapps actually do place demands on the language. I worked on one that had very complex transactions and involved a lot of math. And this was not any kind of scientific app and the company was not scientifically oriented either. I don't think it's fair or accurate to write these apps off as a 1% minority.

0
Comment actions Permalink

Marc

>I think it has to be about Ruby as well.
>

Technically, yes. Practically, it wouldn't make sense to ask JetBrains
to assign resources to develop a Ruby-editor plugin. There is no real
demand for that.

A Rails-editor plugin, on the other hand, is muchly needed, as
- there is none on the market, and
- the number of developer is exploding.

Alain

0
Comment actions Permalink

Rob Harwood wrote:

It is possible, but I strongly doubt it. Rails has 'crossed the
chasm', being used on real projects by real people, Django is still
in the innovator stage. This gives Rails a big head start. People by


If I remember correctly, Django was developed for the Lawrence
Journal-World website (http://www.ljworld.com/) before it was made
public. So, it has been used on a real project by real people. There's
probably more that I'm not aware of, since I'm not heavily involved.

the thousands will start using Rails for real projects in the next
year. Django will be lucky to get to version 1.0. The First Mover


I'm not trying to pick a fight here, but you're very dismissive of a
project that has great potential. They don't seem to be utterly
dependent on community support to move forward since the original and
still current developers are using it do commercial work which no doubt
supports further development. But they do have community support
regardless. Some rather smart people are working with Django and they've
made some nice improvements over the last few months.

effect will be very strong here. Rails has a book, several articles
and tutorials, a large community, etc. etc. etc. The momentum is very
strong. I doubt Django will be much but a blip on the radar for a
long time.


And maybe some people like me just like Python and dislike Ruby because
of it's unfortunate resemblance to Perl.

Now, I'm not saying that Jetbrains should drop everything and add Django
support to IDEA, but to completely dismiss it seems somewhat
short-sighted to me. Keep an eye on it, you may be surprised.

Ciao,
Gordon

--
Gordon Tyler (Software Developer)
Quest Software <http://www.quest.com/>
260 King Street East, Toronto, Ontario M5A 4L5, Canada
Voice: (416) 933-5046 | Fax: (416) 933-5001

0
Comment actions Permalink

Marc Stock wrote:

You are vastly oversimplifying things. ... I don't think it's fair or accurate to write these apps off as a 1% minority.


Not to start an argument, but I think you're vastly underestimating the
number of simple PHP (and even JSP, whatever) webapps that people do,
where performance really is not an issue. Sure, maybe it's not 1%, but I
doubt it's anywhere close to 10%.

--
Rob Harwood
Software Developer
JetBrains Inc.
http://www.jetbrains.com
"Develop with pleasure!"

0
Comment actions Permalink

Gordon Tyler wrote:

Now, I'm not saying that Jetbrains should drop everything and add Django
support to IDEA, but to completely dismiss it seems somewhat
short-sighted to me. Keep an eye on it, you may be surprised.


I'm not completely dismissing it. I'm just saying that compared to
Rails, it is far far behind and has a long way to catch up. And even if
it does catch up (and even surpass; it may be fantastic in 1 year),
Rails will already have drawn a lot of people to it, and may very well
steal Django's thunder.

--
Rob Harwood
Software Developer
JetBrains Inc.
http://www.jetbrains.com
"Develop with pleasure!"

0
Comment actions Permalink

The impression I get is that people keep comparing Ruby to Java, not Ruby to PHP. If you're talking about Ruby replacing something like PHP, it's an entirely different ballgame. PHP, and scripting languages like it are a dime a dozen. Ruby supporters seem to be all over the map on where it fits but it usually seems that they think it fits almost everywhere until you point out a flaw and then it all of a sudden they get a lot more specific.

0
Comment actions Permalink

Rob Harwood (JetBrains) wrote:

I'm not completely dismissing it. I'm just saying that compared to
Rails, it is far far behind and has a long way to catch up. And even if
it does catch up (and even surpass; it may be fantastic in 1 year),
Rails will already have drawn a lot of people to it, and may very well
steal Django's thunder.


Whether it's far far behind Rails, I'm not qualified to say since I've
only used Django superficially and have never used Rails (read about it
a bit though). But as you say, Rails does have a headstart in terms of
mindshare, so it may turn out that way.

I'll just continue rooting for the underdog ;)

Ciao,
Gordon

--
Gordon Tyler (Software Developer)
Quest Software <http://www.quest.com/>
260 King Street East, Toronto, Ontario M5A 4L5, Canada
Voice: (416) 933-5046 | Fax: (416) 933-5001

0
Comment actions Permalink

The impression I get is that people keep comparing
Ruby to Java, not Ruby to PHP. If you're talking
about Ruby replacing something like PHP, it's an
entirely different ballgame. PHP, and scripting
languages like it are a dime a dozen. Ruby
supporters seem to be all over the map on where it
fits but it usually seems that they think it fits
almost everywhere until you point out a flaw and then
it all of a sudden they get a lot more specific.


A recent survey conducted by the Burton Group of current Rails developers (http://weblog.rubyonrails.com/archives/2005/09/23/migrating-to-rails-half-from-php-one-third-from-java) found half came from PHP and one third from Java. This is why Rails is such a disruptive force; it attracts people from PHP, who are looking for a way to do what they currently do, only better; and it attracts people from Java, who are looking for a different way (easier, faster) to do simpler webapps. This is a classic disruptive innovation (see The Innovator's Dilemma; http://www.amazon.com/exec/obidos/tg/detail/-/0060521996). It is disruptive to J2EE, and sustaining to PHP. So basically, PHP will fuel its popularity in the short- to medium-term, but eventually it will become good enough that large numbers of J2EE developers will switch to it in the long-term. You can compare it to the situation of the advent of the personal computer and the spreadsheet. Before that, you had people doing calculations with pen and paper (PHP) and people who used mainframes/workstations to do large calculations (J2EE). The personal computer alone (Ruby) was interesting, but not totally compelling. Adding a spreadsheet program (Rails) suddenly made the personal computer more interesting.

The pen and paper crowd, who couldn't afford, or didn't have the need for, a complicated mainframe or workstation, started buying personal computers to use the spreadsheet. For them this was an improvement on pen and paper calculations.

The mainframe/workstation crowd were split. Most had no use for the inferior personal computer. But there were some who found the relative inexpensiveness and simplicity of the personal computer to be well suited to some of their simpler projects.

Fueled by the pen and paper crowd, for whom the personal computer + spreadsheet was in almost all ways superior, the personal computer became more popular and new sustaining innovations made it more and more capable. Soon, the PC became so good that there was no real need for mainframes and workstations except in some small niche markets.

The future of web app tools will follow a similar pattern. Rails will become more and more popular, fueled by the PHP crowd, and it will become more and more capable. Following the current trajectory, at some point, it will be good enough that most J2EE developers will switch to it, and J2EE will become a niche in the overall market.

0
Comment actions Permalink

Rob

Have you had any success at pushing the idea of a Rails' plugin to other
JetBrainers?
It would be way easier for IDEA to understand Ruby&Rails than for the
few existing Ruby editors to start offering refactorings.

Alain

0
Comment actions Permalink

Alain Ravet wrote:

Rob

Have you had any success at pushing the idea of a Rails' plugin to other
JetBrainers?
It would be way easier for IDEA to understand Ruby&Rails than for the
few existing Ruby editors to start offering refactorings.

Alain


I've talked with Sergey Dmitriev, and sent a Mind Manager map to Maxim
and the rest of the IDEA team. Sergey was mildly skeptical but said he'd
think it over; he had ideas on how it could be done relatively quickly.
I haven't heard anything from anyone on the IDEA team. Not being in St.
P. kind of isolates me from them. They discuss their plans over there,
so I have no idea what they think of it.

--
Rob Harwood
Software Developer
JetBrains Inc.
http://www.jetbrains.com
"Develop with pleasure!"

0
Comment actions Permalink

Rob Harwood (JetBrains) wrote:

>

I've talked with Sergey Dmitriev, and..
They discuss their plans over there, so I have no idea what they think
of it.




I'll cross my fingers. The 20 of them.
Btw, I read yesterday that 4700 copies of the Rails book have been sold
by Amazon during the last week. A sign not to ignore.


Alain

0
Comment actions Permalink

Some links regarding the scalability of Rails apps...

http://raibledesigns.com/page/rd?entry=scaling_with_rails

http://www.loudthinking.com/arc/000479.html

And here's Dave Thomas with some concrete, specific criteria on when and when not to use Rails...

http://blogs.pragprog.com/cgi-bin/pragdave.cgi/Tech/Ruby/IsRailsReadyForPrimeTime.html


I'm somewhat surprised by some of the attitudes on this forum about Ruby/Rails. You'd think that people who "get" why Idea is better than Eclipse and the other IDEs out there would at least be open to, if not supportive of a technology that is designed from the ground up with the same goals (making development faster and more productive).


0
Comment actions Permalink

Hi Marc; hope you don't mind if I interrupt, but here's another way to look at this:

I just want the ruby/rails hype to be separated from the truth. I'm sure it's great stuff but I continually see geeks overstating the situation and I feel it needs to be reigned in a bit[/quote]

JetBrains is a lot like Apple Computers; if they want to stay in business, they need to:

1/. Be ready for the next big thing
2/. Create the next big thing

I agree that you have to watch for the hype. If I spent time learning every framework that 'was the future of web development' then I'd be tired and broke. I kept my eye on Java and that's done fine by me.

The problem is that it is no longer enough for JetBrains. Java is pretty much mature and proven, with an over abundance of tools and frameworks. Free tools and frameworks. NetBeans and Eclipse are progressing at a phenomenal rate, and there may come a time when JetBrains looks at IDEA and realises that they're only adding buggy bloat rather than anything truly useful.

I'm not saying that they will abandon IDEA, but they may need something else to sell along side it. I'm a big fan of Fabrique, but after so long in development, I can't really see it getting much traction when (if) it's released. Jumping on an immature framework with a growing user base could give JetBrains and RoR a boost.

I have no idea whether Ruby will be successful or not. As far as I can tell, it certainly doesn't have the frameworks in place yet. But there is still time.

What I find a bit odd, is that folk like to compare it to J2EE when they spout off the utterly meaningless LoC metric. As if any Java developer is forced to use the whole J2EE stack to build a simple web app. You can build an MVC app with pure JSP if you want to.


0
Comment actions Permalink

It reminds of a program I saw last year, abount an alpinist who was
climbing ultra-light: light-shoes, a little water and a light backpack
with just a few powerbars. He didn't need the tent, the kilos of food,
the intermediary camp, the support team because he was running, and
going up and down in 1 day. That's a real story.[/quote]

The real story is what happens to him when an avalanche hits.


0
Comment actions Permalink

+
A technology that lets you create and deploy a web-app in 1-2 hours...
+

so, they actually don't need any IDE..
if they need one they probably can make one within a few hours/days..
-m

0
Comment actions Permalink

Possibly a controversial opinion here considering the obvious passion of both the Ruby on Rails and the J2EE crowd here but at the risk of getting tomatillos thrown at me here's my opinion...

IntelliJ IDEA is a JAVA development tool and I would suggest it should remain that way. It should support ONLY Java based technologies and their related requirements so things like HTML, JavaScript, CSS, JSF, JSP, Servlets and other Java frameworks (things used by Java developers) should be the only things JB look at when mapping an IDEA roadmap.

With that said, I have nothing against Ruby or Rails and would possibly even start using it myself if JB could develop a good tool for it, and I would support their new venture into the market wholeheartedly. However Ruby has no place in IntelliJ IDEA and neither does Rails since it is a Java technology development environment. If JB are looking at conquering the Ruby market then they should do it in such a way that it is a product in its own right rather than part of any IDEA roadmap.

I would have no problems with them developing a commercial Rails plugin for IDEA so long as it has its own team and does not in any way impede on the progress needed to be made on Java technologies. I would then suggest that the Rails plugin (if successful) should be further developed into an IDE in its own right... IntelliR IDEA anyone?

IMHO very few people will ever use both Rails and Java in the same applications as they are not really complimentary technologies so I see little point in any kind of Rails development being focused within the standard IDEA product.

Also with regards to Rails taking over from Java technologies, I personally cannot see it. There is very little industry support for the product and unless companies like IBM and Sun lead the way in replacing Java with it themselves it is unlikely to attract that audience. It is more likely to attract Perl or PHP and RAD developers than those creating enterprise technology where Java is mostly focused. This is not a slight on Rails or anything like that, just IMO the reality that so long as the big boys still focus their efforts on developing Java technology it will remain prominent.

0
Comment actions Permalink

+ Possibly a controversial opinion here considering the obvious passion of both the Ruby on Rails and the J2EE crowd here but at the risk of getting tomatillos thrown at me here's my opinion...+

All sounds perfectly reasonable to me.


0
Comment actions Permalink

Rob Bradley wrote:

>IntelliJ IDEA is a JAVA development tool ..

I see that as less and less true: css, xml, html, MPS, Visual Fabrique,
the Groovy plugin, ..
IDEA is - also - a language plugin platform, and that's a good thing for
JetBrains: they could have an edge on the Komodo, RailsRad, etc... IF
they don't wait, and IF they take the Rails plugin/IDE development in
their own hands (rather than wait for the community to write a
half-baked one).
There is a real business opportunity, and there is a mounting crowd of
developers who want to develop Rails with - more - pleasure.
Only idiots shant "Hype Hype Hype" when all they know about Rails is the
todo list example and the 15 minute video.


> IMHO very few people will ever use both Rails and Java in the same
> applications as they are not really complimentary technologies

IMO, many developers will just dump Java as a web-platform if their
projects allow it. You may say this is an opinion. I think this is a
fact, but I don't have the numbers to prove it. Yet.

Alain

0
Comment actions Permalink

There is a real business opportunity, and there is a mounting crowd of developers who want to develop Rails with - more - pleasure. Only idiots shant "Hype Hype Hype" when all they know about Rails is the todo list example and the 15 minute video. ]]>

"Idiot"

as in

"anyone who doesn't agree with me"

JetBrains needs to be careful for two reasons:

'Mounting crowds' are sometimes made up of a deceptively small number of people who just shout very loudly.

'Mounting Crowds' often use tools often use languages because they are free; and they also expect their tools to be free as well.


0
Comment actions Permalink

Alain Ravet wrote:

&gt; I see that as less and less true: css, xml, html,
&gt; MPS, Visual Fabrique, the Groovy plugin, ..

I don't disagree with what you are saying but all the things you have listed have one thing in common - they are all Java technologies and are complimentary to what IDEA has been designed to do. Rails is not a Java technology and it is not complimentary to Java, thus there is no need for it in IDEA.

As I said I am not at all opposed to Ruby or Rails, in fact quite the contrary as I can see its possibilities and its commercial viability, however I do not see IDEA as the right place for it. I do believe that JetBrains are the right company to develop it, but not as part of IDEA.

Alain Ravet wrote:

&gt; IMO, many developers will just dump Java as a
&gt; web-platform if their projects allow it.

I disagree. I believe that Java will always be around as a web-platform so long as the big boys support it. Rails can become very popular, possibly even as popular as Java for the web, however it will not be completely dumped so long as the big boys like Sun and Big Blue continue to push it and so long as Microsoft continue to go their own way. If either IBM or Microsoft start pushing Ruby/Rails then it is a very dangerous competitor and could steal the market, unless that happens both Java and .NET will always remain stronger alternatives.

Alain Ravet wrote:

&gt; You may say this is an opinion. I think this is a
&gt; fact, but I don't have the numbers to prove it.

Also hopefully not starting an argument here since I do respect your opinions and don't necessarily disagree with the possibilities of what you are saying, but technically speaking nothing can ever be a fact unless it can be proven. You ARE quoting your opinion, which is fine 'cause I am also giving you mine. I don't claim to posses the facts, but I do believe I have a logical argument that follows historical patterns.

0
Comment actions Permalink

"Idiot"

>as in

"anyone who doesn't agree with me"

>

>

No Rayz, idiots as in :
"All those people saying Rails is great, and __I__ first heard about it
1 week ago. They can't be right. It must be hype, ... till I say it
isn't ... but I don't have the time to learn it yet, so it will be hype
for another 6-12 months. Amen."

Pathetically enough, there's a second kind of idiots that partially
justifies the idiots above reaction. They live in a few dozens empty
blogs, and recently wrote an article that reads like:
"I tried Rails yesterday, after my friend John told me it was so great
and I had to give it a try, and I tried, and I must say it is great.
It's all I have to write right now, but I'll tell you more later,
because I don't have the time right now, and I don't it very well yet,
as I just started yesterday."

Alain

0
Comment actions Permalink

+No Rayz, idiots as in :
"All those people saying Rails is great, and __I__ first heard about it
1 week ago. They can't be right. It must be hype, ... till I say it
isn't ... but I don't have the time to learn it yet, so it will be hype
for another 6-12 months. Amen."+

Yep. Basically anyone who doesn't agree with you.

+Pathetically enough, there's a second kind of idiots that partially
justifies the idiots above reaction. They live in a few dozens empty
blogs, and recently wrote an article that reads like:
"I tried Rails yesterday, after my friend John told me it was so great
and I had to give it a try, and I tried, and I must say it is great.
It's all I have to write right now, but I'll tell you more later,
because I don't have the time right now, and I don't it very well yet,
as I just started yesterday."+

Yes, this kind of fanatical ranting is what JetBrains has to be careful of. Alot of screaming, but no evidence.

Now do you have any proof of this mass takeup, apart from "I love it!! So it must be a hit!"?

0
Comment actions Permalink

Rob

>> Have you had any success at pushing the idea of a Rails' plugin to
>> other JetBrainers?
>
>

I've talked with Sergey Dmitriev, and ...
Rob





Any success? Any progress? I'd hate having to spend 295$ on
ActiveState's Komodo "just" to get Ruby syntax checking, code
completion, code highlighting, svn, etc, etc ...



Alain

0
Comment actions Permalink

Hello Alain,

>>> Have you had any success at pushing the idea of a Rails' plugin to
>>> other JetBrainers?
>>>
>> I've talked with Sergey Dmitriev, and ...
>> Rob
AR> Any success? Any progress? I'd hate having to spend 295$ on
AR> ActiveState's Komodo "just" to get Ruby syntax checking, code
AR> completion, code highlighting, svn, etc, etc ...

Rob no longer works with JetBrains.

We are keeping Ruby/Rails in mind as a potential future product, but we do
not have any plans to start working on it in the near term.

--
Dmitry Jemerov
Software Developer
JetBrains, Inc.
http://www.jetbrains.com
"Develop with pleasure!"


0
Comment actions Permalink

Alain Ravet wrote:

Any success? Any progress? I'd hate having to spend 295$ on
ActiveState's Komodo "just" to get Ruby syntax checking, code
completion, code highlighting, svn, etc, etc ...


I believe there's an Eclipse-based plugin for this stuff, which is free. You
could also develop your own Language Plugin, I think it wouldn't be more than a
few days work to get syntax checking/highlighting and completion. IDEA already
supports SVN.

0
Comment actions Permalink

So you're willing to bet your career on what is essentially an unproven technology? I sure as shit wouldn't. Rails has been out for less than six months so I just don't get how you can make such a bold statement as...

"IMO, many developers will just dump Java as a web-platform if their projects allow it. You may say this is an opinion. I think this is a fact, but I don't have the numbers to prove it. Yet."

Thinking it's a fact just shows that you are just as consumed by the hype.

0
Comment actions Permalink

Marc Stock wrote:

So you're willing to bet your career on what is essentially an
unproven technology? I sure as shit wouldn't. Rails has been out
for less than six months so I just don't get how you can make such a
bold statement as...


Rails 0.5.0 was released a year and a quarter ago on July 24 2004.

0
Comment actions Permalink

Dmitry Jemerov (JetBrains) wrote on 04/11/05 00:27:

Rob no longer works with JetBrains.


That's a shame :( Whats he doing now? Hopefully something exciting and
fullfulling.

0
Comment actions Permalink

Anything prior to 1.0 is either considered "beta" software or not feature complete. It is still unproven software (unless you have very weak standards). Again, would you bet your career on it?

0

Please sign in to leave a comment.