why companies don't use in HaXe

[michael solomon]

hi,
First, this is my first post here so.. my name is Michael. 22 years old from israel and i'm apologize if my english little funny sometimes - i'm working on it :)
i heard about Haxe 3 years ago from my friend and then i said to myself interesting if it's working well and continue to grow because if it does it's will be awesome so i waited few years(meanwhile i completed my duty in IDF).
Now i becomes interested in Haxe developing again but although that Haxe continued to grow and add a significant features i still don't see companies that going to develop in Haxe and i don't see plenty of Haxe job requirements.
I thought that at least(at the Min of the Min...) Haxe can be incredible replacement for Flase/AS3 but i don't see this happen.
What to heck stops Haxe to grow to the enormous size that she should be?
p.s: i am very enjoy to develop in Haxe!!! it's cool language AND Open Source :)
p.s2:i don't try to say that Haxe is failure. I REALY think that Haxe have a good chance to be what she strive to be and it's freak me out that although all amazing features Haxe still don't have a good grasp in the market.
Regards,
Michael

[michael solomon]

lol... its impossible to edit post in google group so i correct this line in my post:

I thought that at least(at the Min of the Min...) Haxe will be incredible replacement for Flase/AS3 but i don't see this happen.

[dlots]

Most of the giant languages  come from persons associated in some way to have the communal authority to declare a new paradigm for some set of functionality or from some other popular authority to declare a new iteration in language. Eg:

original language designers associated with academia with a declaration to provide unification for a set of functionality
a major standardizing initiative
a major corporation

[TroyWorks]

>What to heck stops Haxe to grow to the enormous size that she should be?

Lots of reasons:

Part: Give it time, the haxe foundation is making progress in the area. Rome wasn't build in a decade.

Part: It's sad but the success of anything has little to do with how good or useful it is...millions of patents and good businesses never had mainstream commerical success .   This problem is not unique to Haxe, others ask it about Lisp and Rebol, Haskell, Scala etc and have been for decades.

Flash success came from massive money.  Tens if not hundreds of millions of dollars over 2 decades in Macromedia then Adobe spending massive money in advertising and events, selling pricey tools, marketing and corporate clients with amazing showcase pieces, and at the time the lack of html to do anyting comparable.

Today, Haxe has to compete with seemingly hundreds of different competitors (unity, swift) that have clearer objectives and stronger corporate support...basically they all talk louder in a crowded room.   At surface level, a engineer manager says ..oh it's good enough for facebook, I want to be like them.  Facebook's React has Facebook and Instagram as champions. Twitter and Google have their own built in house.  Who with gazillions of dollars is championing and champion haxe?  Tivo, Prezi, Nickolodean use it but they are likely orders of magnitude smaller than twitter, google, facebook etc.  As it's open source there is no strong motivation for big corps to want to control something they can't.

[Dario Elyasy]

Hi michael, it is good to see other Israelis getting excited about Haxe

I would like to add a few comments to the what TroyWorks wrote.

Although most of what he wrote is true, I don't agree that you must be a big corporation to promote successfully a new language. See the example of the relative failure Google Dart, especially when compared to Google Go that is instead a quite a success in term of adoption.

A very important component in the success of a new language is if it has a compelling use case, a problem that it solves better than other languages. See the example of CoffeScript, the most successful language that transcompile to Javascript. This is an open source community backed language like Haxe. The main use case it solved is to give Pythonists and Rubyist a smooth way to move to the rising Javascript platform for backend development (but also frontend).  See also node.js. This is also a open source community project. The problem it solved: allow backend development in Javascript and code share between front-end and back-end and also allow for DIRTy web applications (data intensive real time applications), Node.js has been actually adopted by big corporation like ebay, paypal, yahoo google, and many others (see http://nodejs.org/industry/).

I think that Haxe has several use-cases where it has winning cards vs other languages. 

Personally I think that for example Haxe can be an ideal tool for startups: startups have very limited resources and can't allow themselves to have separate teams taking care of different aspect of the product (front-end, back-end, mobile, etc.). This is particularly true today with the explosion of the number of platforms that a new software product can target (smartphones tablets various PC OSs , consoles, wearable devices, etc.), A possible alternative is HTML5/Javascript, but Haxe/Js is much better than plain javascript. 

The problem with Haxe for startups is tooling and documentation: in a startup you cannot loose time because you don't have the usual modern tools for fast development (a good integrating ide with refactoring capabilities, good debugging tools and so on) and also  

lacking documentation is a big problem, because I cannot loose time because of that, if all the point of using haxe is fast development for multiple platforms. Also it makes much more difficult to recruiting and training new people for the startup, which is already a challenge..

Dario

On Thursday, September 18, 2014 7:30:50 AM UTC+3, TroyWorks wrote:

>What to heck stops Haxe to grow to the enormous size that she should be?

Lots of reasons:

Part: It's sad but the success of anything has little to do with how good or useful it is...millions of patents and good businesses never had mainstream commerical success .   This problem is not unique to Haxe, others ask it about Lisp and Rebol, Haskell, Scala etc and have been for decades.

[Bernd Ritter]

I wouldn't say that. I came as a private person to haxe and started to love it. Now at my workplace we use alot of open source software and I mentioned haxe's multiplattform possibilities, especially for andoird. So the chances are high that we might use haxe with our next android project.

But most companies stick to what they already know.

--
To post to this group haxe...@googlegroups.com
http://groups.google.com/group/haxelang?hl=en
---
You received this message because you are subscribed to the Google Groups "Haxe" group.
For more options, visit https://groups.google.com/d/optout.

[Philippe Elsass]

Haxe wouldn't be good choice for a startup, at least if you're here for the money: huge growth, quick sellout. You want the most popular technological stack.

Coffeescript got some popularity thanks to the RoR hype.

Dart is still very young but it is actively investigated by many companies. Probably a lot more than Haxe already because it's Google-backed.

What do we have in the companies that are using Haxe?
- indie developers: "this game engine looks good; oh the language is Haxe"
- Nickelodeon pays contractors making web games; they were convinced by the benefits of using Flambe (efficient html5+Flash output) and they can afford encouraging (requiring?) contractors to use it,
- established companies (TiVo, Prezi, Massive Interactive): "we investigated this Haxe language and we think, despite the risks, that it's worth investing our time and money building/porting to it and training our engineers."

At Massive Interactive, the last 4 hires where: 2 had never heard of Haxe, 1 had played a little bit a long time ago, and 1 had a some experience doing side projects in Haxe.

I'll say it again: we must invest the JS world. Make it super easy and well documented to use Haxe using the JS toolchain (gulp/grunt/yeoman/node/...). The benefit has to be immediate ("don't make me think").

[Dan Korostelev]

A huge +1 to JS world opinion. I spent some time improving stuff for externs (@:jsRequire, haxe.Rest), but the biggest thing we want is externs themselves and I think we really need officially supported quality externs for more popular libs such as nodejs.

четверг, 18 сентября 2014 г., 11:59:37 UTC+4 пользователь Philippe Elsass написал:

[Dario Elyasy]

Hi Philippe

Can you elaborate a little more about the reason why you don't thing Haxe a good choice for a startup?

[Mark]

ergh. Shalom then. "שָׁלוֹם" if google REALLY speaks Hebrew, not just imitating! :D

So, on top of what others say, my 'extension' is that one has to think differently if leading a big(ger than few heads indie) organization with huge business:
One has to take into account the market risks, and those become the most important thing. "Quantity over quality", so to speak.


Here's the long version if interested. Highlighted keywords for med-long version :P Enjoy! ;)

1. A "backed language", having a huge corporation behind is less likely to disappear one day to another, as well as more likely to stay up-to-date and follow/pick up new developments (e.g. hardware or OS), fix security holes (and spread the fix), also has a bigger "tester for free" (user) base, etc.

On the other hand, if Nicolas loses interest in HaXe (or something even worse happens), that's a HUGE drawback for the language, if not the end of it, i.e. slowly fading from the "market" as more and more devs lose interest, who were there just to work with Nicolas, the father of the language (no offense or anything like that here!).
See? A single person! One guy! For Java / dot NOT (hehe) / etc, or even for overhyped ones like Unity -ie stuff with better advertising than actual capabilities-, tens, or hundreds, or even thousands have to "disappear" for anything similar to happen. Which has a mathematical chance of course, but technically is not possible to happen.
Another option is a huge market loss (also low chance), like, billions of dollars loss with nobody around, who is willing to buy up the 'language' on 0.0000001% price.

EXAMPLE: Look at (e.g.)Apple. Steve Jobs died. From business perspective, nobody cares. One guy, even if the "head of all" is nothing in a crowd. It's like tears in rain.
Or the 'celebrity leak' - Apple or "cloud computing" may have lost some fans (0.00001%?) but that's all. A small company losing the same amount of (e.g.) 5-10 'backers', and/or disappointing 2.000 persons would be busted and gone forever the next day. RISK. And a spooky one.


2. Employees. You need "headcounts", not people. You need "resources" and "workforce", not "good guys to work with". You need everything you ever found on bullsh.t bingo in the "corporate world" and in huge numbers! If you go out to the job market to find (e.g.) "java developer", or "iOS/Android developer" - you throw a rock and it bumps on the head of 'one of those dudes'.
If you look for a HaXe developer, good luck hiring let's say 10 in your region (or if "The Board" is putting pressure on you, to find 10.000 in two months...).
Same applies to replacements (sick leave, imprisonment -it happens dude, it happens! :)-, birth, death, new workplace, productivity drops as employee gets unhappy with whatever, need to promote so starts managing, i.e. not working anymore, etc, etc... endless list... The List Of Terror with unique guys on board, in fact).


3. Salaries and finances! It is very close to #1 and #2, pretty much summing them up but still I hope you find it an interesting read:

a) Crowd-people are cheaper. Unique people are more expensive. Cost 'x' gives 'n' mandays for HaXe, 2*n or 5*n for a "crowd language"
b) Nobody cares about quality, but "deadlines" and "deliverables" only. Cheaper resource is less productive, provides lower quality code, but it does not matter. App. was done fast, it's working(ish) and is on the market. Buggy? Will be fixed! Slow? Will be faster! Promise! Never happens, though. But hey, here's a useless but shiny new feature in HD resolution! (see the irony, right? :)
c) The satisfaction of The Holy Board! (BOW DOWN NOW!!! NOW!!! :) You see. When you present what you achieved that Q or that Y, you show numbers and numbers only. Customer satisfaction is not on that graph (even though they might be happy as if they are other "corporate world companies" they also bother with numbers only).
Nobody cares about anything else. See bullets 'a' and 'b': The crowd of cheap did stuff fast. Small numbers out (costs), big numbers in (profit). 'X' new heads hired, 'Y' left (fluctuation). 'Y' < 'X' == growth == Everybody is happy.


Summing up, breaching out and hitting into the "corporate world" is very difficult without huge hype and crazy amount of money, as well as without evidence that if half of your company gets wiped out tomorrow (super virus? frenzied space gorilla? a Board decision? nah. super virus!), the show still goes on.


Hope you enjoyed reading ;)
Mark

[Jason O'Neil]

On Thu, Sep 18, 2014 at 8:34 PM, Mark <markk...@gmail.com> wrote:

1. A "backed language", having a huge corporation behind is less likely to disappear one day to another, as well as more likely to stay up-to-date and follow/pick up new developments (e.g. hardware or OS), fix security holes (and spread the fix), also has a bigger "tester for free" (user) base, etc.

On the other hand, if Nicolas loses interest in HaXe (or something even worse happens), that's a HUGE drawback for the language, if not the end of it, i.e. slowly fading from the "market" as more and more devs lose interest, who were there just to work with Nicolas, the father of the language (no offense or anything like that here!).
See? A single person! One guy! For Java / dot NOT (hehe) / etc, or even for overhyped ones like Unity -ie stuff with better advertising than actual capabilities-, tens, or hundreds, or even thousands have to "disappear" for anything similar to happen. Which has a mathematical chance of course, but technically is not possible to happen.

Can we end the myth that Nicolas is the only person working on Haxe?  He's still in charge and he contributed the most over time, but these days he's not even the main developer:

https://github.com/HaxeFoundation/haxe/graphs/contributors?from=2011-12-24&to=2014-09-18&type=c

Simon, Dan, Caue, Andi, Heinz, Hugh and others are making far more regular contributions.  It's a small list but it's diverse enough that any one person leaving won't end the project forever.

Also, I don't buy the "A big corporation means the language won't disappear" argument.  What happened to Visual Basic 6 developers?  What happened to ActionScript 3 developers?  Etc.  A big company can give up on a platform at any moment.  What makes a platform last is multiple companies using it, so that if any single group stops, other groups still have an interest in maintaining it.  I think Haxe is at the point now where if Nicolas made a hit game and earned millions of euros and retired to the Bahamas and never opened his laptop again, there are enough developers and companies using Haxe to keep it going without him.  (Not to mention, maybe then we'd get short lambdas!).  If Google decides Dart isn't interesting anymore, Dart ends that day.  Ouch.

Languages / Frameworks like C, Java, Javascript, Python, NodeJS and Haxe are used by many companies, and maintained by many companies, and in my opinion are more robust and trustworthy than a language controlled and maintained by one company.

Jason

[Philippe Elsass]

I guess that would depend on the startup, but there would be a number of pragmatic considerations if you want to become big quickly:

- is the technology stack known to scale well and are experienced specialists available on the market?

- will it be easy to grow the engineering team really quickly?

We love Haxe but use it only for the front-end and it's hard enough to recruit people.

For the back-end (CMS/AWS) we use .NET (could have been PHP or Java) as that's not an aspect of the stack we're willing to take risks.

--
To post to this group haxe...@googlegroups.com
http://groups.google.com/group/haxelang?hl=en
---
You received this message because you are subscribed to the Google Groups "Haxe" group.
For more options, visit https://groups.google.com/d/optout.

--
Philippe

[Philippe Elsass]

+1 Jason, another myth that needs to be cleared

--
To post to this group haxe...@googlegroups.com
http://groups.google.com/group/haxelang?hl=en
---
You received this message because you are subscribed to the Google Groups "Haxe" group.
For more options, visit https://groups.google.com/d/optout.

--
Philippe

[Mark]

I initially didn't want to answer but the +1 made me :)
I never said HaXe is a "one man army" development. Read what I wrote... No myth or mistification there. On the other hand, I understand you don't want this myth hanging around (if it does) so no offense taken (hopefully, also not given!).

Related to:

> "Also, I don't buy the "A big corporation means the language won't disappear" argument. What happened to Visual Basic 6 developers?"

Nice question. Google for an answer for interesting results. Crazy, really :D

SO! if something disappears, what happens to people doing that is easy to answer. Like. "What happened to ...": steam-engine train drivers, sword/armor/etc blacksmiths, hangmen and other executioners & torture masters, vassals or Waldo. If we put aside the obvious answer (which is, nobody cares), the standard protocol was followed I assume:
The jobs are gone so the proficiency become useless --> one has to learn a new proficiency or starve to death (or, like, for torture masters, find cleverly the market gap and get into tax inspection. same boat, different flag).

Related to companies, VB did not disappear one day to another. It has become discontinued, not removed like fappening links from google.
It was announced, time was given to adopt. Microsoft was and is not in a position to brush away something off the table they don't like. The reason is, they would lose market. Google is in the same boat. As well as the rest of the big players (yes, there are some examples but those are products considered vaporwares -mostly?- or a market move considered idiotic. like. I don't know. the desktop gallery for win2K?).

Oh! Did I mention these over-sized monster companies have strict lifecycle product policies? A quick search resulted in:
Microsoft: http://support.microsoft.com/lifecycle/?LN=en-us
Apple:       ftp://ftp.attglobal.net/pub/client/apple/mac/product_lifecycle_support_policy.pdf
Google:     Now here are MANY! An example: http://www.google.com/patents/US7890626

And rest assured, grotesque giants follow their policies. Always. At.all.cost.


The thing is, yes, a 'corp world' company has the power to remove something from the table next day. I agree and admit it. True.
The question is, do they dare?
If google does it with Dart, nothing's on, right? What if google comes out 2 weeks later with e.g. "The BullsEye" as a successor of Dart? Then flushes it down?

2-3 moves like that, and nobody will be interested in any corp.action trying to win dev/e-mail/social media/whatever market on their chesstable of 'market sectors' and 'business area'.
E.g. Microsoft bought up Mojang AB for an insane 2.5 billion USD? Do you know why? Oculus Rift (fb)? Beats (Apple)? They are like girls ("just wanna have fun, oh-uh-oh-uh!") and fun for them is, winning market. With a silly, endless resource called "money". I really hate I have to work for it!


Cheers,
Mark

[Daniel Glazman]

On 18/09/2014 00:44, michael solomon wrote:

> What to heck stops Haxe to grow to the enormous size that she should be?

I asked myself the same question a while ago. It took me years and years
to discover Haxe.

That's essentially because of marketing. Marketing
a programming language like Haxe requires attending conferences to
present and demo it. And not only in the gaming field. Implications are
simple: travel budget, exhibition booth costs, time and energy to
produce talks, demos and tutos, etc. Evangelism made by remarkable
evangelists is a key. It also requires making press releases that some
key tech journalists will relay.
(speaking of that, I would recommend a "Made with Haxe" logo and some
swag early adopters could wear/show at events: t-shirts, bags, etc.
Start a store at SpreadShirt?)

See Apple's Swift. Even if it aims at OSX and iOS only, it shares some
goals with Haxe. But it has the enormous marketing weight of One
Infinite Loop at Curpertino behind it. Yes, I know, it also has the
full power of Cocoa, and the zillions Obj-C developers as early
adopters. But still, w/o marketing and docs, it will go nowhere.

The other factors are well known:

- good debugging environment, whatever the target
- stable (and possibly expanding) core and targets team being paid for
their work on Haxe on all platforms
- advocates (that's a bit different from evangelism)
- excellent, complete and up-to-date documentation

</Daniel>

[Nicolas Cannasse]

> p.s2:i don't try to say that Haxe is failure. I REALY think that Haxe
> have a good chance to be what she strive to be and it's freak me out
> that although all amazing features Haxe still don't have a good grasp in
> the market.

I wouldn't say that "companies don't use Haxe", since there are already
many companies having large teams using Haxe on a daily basis.

To quote a few (that I know about):

Nickelodeon, Disney, TiVo, Prezi, Motion-Twin, GameDuell, Massive
Interactive, Stencyl, etc et etc

So I would guess your question is why there is not a LOT MORE companies
using Haxe ?

I think there are a few reasons:

a) we are not backed by a big company name, which means we need to do
more marketing to get the word out, Haxe Foundation is working on
improving this. OTOH this is also why we will still be there in 10
years, compared to a big company that can entirely stop developing a
tech on a whim (numerous examples exists, Flash being one of them)

b) Haxe is not easy to understand for many people. Not the language, but
what the technology brings. We don't have an ideal "get started"
tutorial, since there are many many many things you can do with Haxe.

We need more dedicated high level technology "built with Haxe" that can
teach people for a specific given usage. HaxeFlixel.com is a good
example of that: it focus on game developers, provides samples and
tutorials for this specific audience.

We need the same for the various areas where Haxe can be used. And I
don't think that having all of them on haxe.org will help a lot, would
only makes things more confusing.

Best,
Nicolas

[Tarwin Stroh-Spijer]

As a small aside, whenever I mention that I use Haxe to random people at the pub (Silicon Valley so they're all engineers) they all have heard of it, and are pretty interested when I explain.

Tarwin Stroh-Spijer
_______________________

phone: +1 650 842 0920

Developer at Fanplayr Inc. (Palo Alto)

Original at Touch My Pixel (touchmypixel.com)
_______________________

[Adam Harte]

You are right about haxeflixel.com. There getting started docs are an amazing entry point to Haxe as well as there library. I point anyone who is interested in Haxe to there website. Once they have a basic understanding for how Haxe works in that context, it blows there mind to see everything else it can do.

[TroyWorks]

> See the example of the relative failure Google Dart, especially when compared to Google Go that is instead a quite a >success in term of adoption.

Still google though and I'd wager that neither Dart or Go would have the visibility without the Google attached to it.  Google is still made of humans so products like Google Wave and others don't make it.

> See also node.js. This is also a open source community project. The problem it solved: allow backend development in Javascript and code share between front-end and back-end and

Right on, though from a CTO hat, the problem it solves is one programmer can do both sides, rather than needing to, javascript it not necessarily the first choice.

 

Personally I think that for example Haxe can be an ideal tool for startups:vs Also it makes much more difficult to recruiting and training new people for the startup, which is already a challenge..

Having started and run several startups, I agree Haxe could be great or horrible depending.   Contrast Facebook's choice of PHP (which is transcoded to C) as the core language, vs selecting Haxe.  PHP developers are a commodity, almost nobody on a job board lists Haxe. Even if Haxe is selected, who to train, what courses books exams exist at what cost, what existing code can be used on github, answers on stackoverflow etc.  So a non-technical team could not find techincal resources, as well a non-technical team is not savy enough to differentiate between the louder trumpets of Unity, Titanium etc...barely having a concept of how projects are built at all.

Another less easy issue is that many coders like the craft of coding by hand..and/or comfortable with repetition/monotony...aka script kiddies copy and paste profits. Who might think ".all that extra type information so much to type ugh bleh."   So the tradeoff of having a higher level with overhead of externs and the like is an immediate non-starter, or marginal increase.  Typescript's sluggishness is probably enough for many to think that all such approaches might be more friction than it's worth and give up anything beyond that..

>JS world. Make it super easy and well documented to use Haxe using the JS toolchain (gulp/grunt/yeoman/node/...). The benefit has to be immediate ("don't make me think").

Agree 100%.  Phillippe.  Glad to hear that at least some companies are hiring in Haxe.  I've been looking in to using gulp with haxe just the other day , as well as merging with React.js/jsx display.

[Dario Elyasy]

Thanks to TroyWorks and Philippe for your insights about the needs of a start-up and pros and cons of using Haxe in such context, they are very helpful. Now about the issue of the lack of trained Haxe developers on the market, if we restrict ourselves to Haxe/JS, what do you think about the option of using TypeScript together with Haxe/JS? Their syntax is very similar, so you can basically hire TypeScript developers and perhaps move codebase if required from Haxe/JS to TypeScript and viceversa according to needs, and/or convert TypeScript developers to Haxe/JS? Do you think that it is reasonable to assume that supply of TypeScript developers is going to increase since  TypeScript has many common traits with the new Ecmascript 6 standard?

Dario 

[Philippe Elsass]

Getting into Haxe isn't a problem in itself, developers we train get very quickly familiar with it. But training means you need to have enough people capable or mentoring the new ones.

 

It's just easier to hire a lot of people for JS using some popular framework instead of an unfamiliar language and unknown frameworks: devs won't feel confident to be candidate, and also many think about building a cool CV full of buzzwords. 

Maybe the approach would be to say: "we hire JS developers to do web development, and more precisely we write code in Haxe JS (think JS with classes)".

On Fri, Sep 19, 2014 at 7:39 AM, Dario Elyasy <dar...@beyond-eye.com> wrote:

Thanks to TroyWorks and Philippe for your insights about the needs of a start-up and pros and cons of using Haxe in such context, they are very helpful. Now about the issue of the lack of trained Haxe developers on the market, if we restrict ourselves to Haxe/JS, what do you think about the option of using TypeScript together with Haxe/JS? Their syntax is very similar, so you can basically hire TypeScript developers and perhaps move codebase if required from Haxe/JS to TypeScript and viceversa according to needs, and/or convert TypeScript developers to Haxe/JS? Do you think that it is reasonable to assume that supply of TypeScript developers is going to increase since  TypeScript has many common traits with the new Ecmascript 6 standard?

Dario 

--

To post to this group haxe...@googlegroups.com
http://groups.google.com/group/haxelang?hl=en
---
You received this message because you are subscribed to the Google Groups "Haxe" group.
For more options, visit https://groups.google.com/d/optout.

--
Philippe

[clemos]

Hi,

Yeah I totally agree with that.

It's much easier to sell Haxe/JS to your boss / client when presenting it as

-  "just JS, but statically typed, and cleaner"

- cleaner / stricter / more flexible than Typescript

- not such a huge commitment, after all (you can follow some good practices that make your code easy to port back to plain JS, or Typescript, or whatever)

- great potential (eventually compile to Cocktail, reuse in nodeJs etc)

Best,

Clément

[Saumya Ray]

I would say, if at all this is true that "many companies not using HAXE", thats a good sign.

HAXE should be used in specific companies. And from that, one can choose whether to work for a company which is evaluating new technologies or for a traditional shop.

Again I personally feel, HAXE should not go to corporates in the way Flash targeted with Flex. 

HAXE is huge, multi personal and can deliver, those who understand will use.

Just my 2 cents.

[Tarwin Stroh-Spijer]

@TroyWorks. Facebook does not compile any more. They wrote their own VM for PHP (HipHopPHP - http://hhvm.com/) which actually runs FASTER in general than their compiled version did. Speaking to the guys who made it they were encouraging about Hacks actually, and wondered how hard it would be to have a "Hacks" target - that is the added extras version of PHP they made.

Tarwin Stroh-Spijer
_______________________

phone: +1 650 842 0920

Developer at Fanplayr Inc. (Palo Alto)

Original at Touch My Pixel (touchmypixel.com)
_______________________

[TroyWorks]

Thanks Tarwin for the correction on Facebook's VM for PHP.  Interesting to hear, though not suprising it runs faster, just in time optimizations based on current load rather than premeditated guesses.    Java's VM has had similar capability since 1999 (!) http://en.wikipedia.org/wiki/HotSpot.

I actually miss that a bit, the ability to modify and debug code on the server *mid request* was really powerful, and uncommon.

[TroyWorks]

On Friday, September 19, 2014 11:30:16 AM UTC-7, Saumya Ray wrote:

I would say, if at all this is true that "many companies not using HAXE", thats a good sign.

I disagree, bigger is better...the benefits we all receive are related to the size of the community.  People who Haxe is too complicated for, will self select out.  People who might love Haxe never hear about it, thus case studies, git libs, improvements to the core languages.  There could be meetups everywhere like their are JS...but there's not.

In some cases it's the people who need to hear most, who don't know, and worse will never know...and those who love Haxe but can't find work in it, have to move on to things that will pay the bills, are significant tragedies to our community....and meanwhile inferior approaches do thrive and worse become mainstream burden on the world, making Haxe that much harder to be found.

How much simpler would web browsers and the web experience be if the fundamental languages prohibited runtime errors, that could be easily detected at compile time?  How many developers hours and testers hours saved globally?  How many 10's of millions on profiling and optimization of various virtual machines (like google V8) could be avoided with stronger typing. 

From an Education perspective, when I was in school for computer science, and playing with computers after school...I'd much rather have learned Haxe with the ability to build reuseable cross platform algorithms, than the dozens of disposable languages I've ended up learning over the years e.g. fortran, pascal, visual basic, etc. Plus the 100's of frameworks and variations in basic libraries. But that will never happen without Haxe reaching critical mass.

[deb...@eclecticdesignstudio.com]

I know lots of companies that use Haxe :) Its slowly building up steam - give it time! 

On Wednesday, September 17, 2014 3:44:15 PM UTC-7, michael solomon wrote:

hi,
First, this is my first post here so.. my name is Michael. 22 years old from israel and i'm apologize if my english little funny sometimes - i'm working on it :)
i heard about Haxe 3 years ago from my friend and then i said to myself interesting if it's working well and continue to grow because if it does it's will be awesome so i waited few years(meanwhile i completed my duty in IDF).
Now i becomes interested in Haxe developing again but although that Haxe continued to grow and add a significant features i still don't see companies that going to develop in Haxe and i don't see plenty of Haxe job requirements.
I thought that at least(at the Min of the Min...) Haxe can be incredible replacement for Flase/AS3 but i don't see this happen.

What to heck stops Haxe to grow to the enormous size that she should be?

p.s: i am very enjoy to develop in Haxe!!! it's cool language AND Open Source :)

p.s2:i don't try to say that Haxe is failure. I REALY think that Haxe have a good chance to be what she strive to be and it's freak me out that although all amazing features Haxe still don't have a good grasp in the market.

Regards,
Michael

[michael solomon]

Ok, first of all, i'm sorry it took to me a long time to reply, i just wanted to read all your's replies and it's took me around 2 hours to do this and i must say that i also very interested to read all the sub-threads that "born" from my question:)
Second, i'm very glad to see that haxe have a very good community from little experience that i have - strong community is very important to open source project.
Third, i gathered all your's thoughts that said here and i hope the following lines conclusion this thread well:
the problems that Haxe stands in front of them:

1.     a lack in tutorials.
2.     relatively poor documentation.
3.     Haxe is complicated in some respects
   
4.     backed with a big company.
5.     a lack in employees.

I think that 4 and 5 it's result from 1&2. who introduced me to Haxe 3 years ago despaird in the installation step! i think that what we need it's to make Haxe more friendly.
in saturday i was a time to think about optional answer to this problems - and based on your's thoughts and my feeling about what missing to learn Haxe better and make her more friendly:

1. future plans... - what Haxe developers working on and what they future plans, it's improve the confidence that Haxe don't gonna shut down so quick :)
   
2. show what we done with haxe - in addition to this page in haxe.org we need to open(maybe in the same page) projects that independent developers did in Haxe.
   
3. build a dedicated page/site for all target - maybe we need to open self site/page to every platform or target or whatever that Haxe can export to, this site built by developer that develop before to this target(in the original target language), it's necessary because maybe one guy that developed before in JS can explain and put in this site/page better examples and stuff for JS developer , he know to explain Haxe to his original community better because he went through the process too.
   
4. books - we must to give an option to Haxe developer to know Haxe on all her depth, we need book like "CLR via C#"... at present we have 2 books and both of them outdated.
   
5. adversment jobs - when i search on Google i dont see any page of Haxe.org or agent from Haxe community that  publish a jobs in Haxe, I know it's there somewhere but if you dont in Google first page you don't exist.
   
6. a - maybe Haxe just need it start to focus in her strong sides and then add more targets? just for exposure.

Again, i'm sorry about my lame english, i hope that you don't suffered too much.
I realy hope that Haxe will grow up and i think that i gonna to give her a true chance, meanwhile i try to spread Haxe to all who i know in Israel (including in .NET meetups :D).
thanks for all responses,
Best,
Miki

[Hugh]

I think before you even get this far, we should consider:
How many companies considered haxe, but rejected it (however quickly) vs those that did not even consider it.

If the first part of this is anything less than 50% (I suspect it is) then fixing documentation is not going to fix very much.

Of course I could be wrong in my estimates, so we need to MEASURE:

1  - What fraction of companies have got as far as browsing the haxe website for some idea of whether it would be useful

   2  - Of these, how many  rejected it

      3  - Why

We are only asking Q3, but Q1 is most crucial,  The people in this discussion have preselected themselves as people who know about haxe.  What about this others.

Hugh

[James Wil]

We don't care about what company use x or y, nowadays people try/use many technology at the same time, it's not good to stay with just one, that's how we learn and that's how we evolve

Haxe is great, especially for indies ! it's free, open source, and crossplatform!

To get more attention Haxe should clearly say what you can do with Haxe, and provide a better doc, but these things takes time to be done properly

[michael solomon]

 I think before you even get this far, we should consider:
How many companies considered haxe, but rejected it (however quickly) vs those that did not even consider it.

If the first part of this is anything less than 50% (I suspect it is) then fixing documentation is not going to fix very much.

Of course I could be wrong in my estimates, so we need to MEASURE:

1  - What fraction of companies have got as far as browsing the haxe website for some idea of whether it would be useful

   2  - Of these, how many  rejected it

      3  - Why

We are only asking Q3, but Q1 is most crucial,  The people in this discussion have preselected themselves as people who know about haxe.  What about this others.

Hugh

lol , I think that we get here the "Chicken - egg" question, who created first ? :)
I think that the exposure of open source project it's first of all from the people and not from the companies.

[Hugh]

Ok, so perhaps the question is "how many developers who are in a position to make technology decisions have visited the haxe website".

But this point is the same - the docs only help once they get to the website.  Should we be working directly on getting these developers to the website, rather than working on documentation.  Obviously both help, but with limited resources,  which one will better fix "why companies don't use in HaXe".

And you can ask a similar question - "How many indie game developers have visited the haxe website".  Probably bigger %, but I still hear "why have I not heard of this before?"

Hugh