Yes, if you have a 'product' perspective, but others will have a service 
provider perspective and would like to see employers committed to Clojure 
and looking to engage with practitioners.+

On Sunday, March 26, 2017 at 4:49:45 AM UTC+11, Dragan Djuric wrote:
>
>
> Isn't it advantageous in some sense to have access to stuff that your 
> competition doesn't have?
>
> On Friday, March 24, 2017 at 11:05:24 PM UTC+1, piast...@gmail.com wrote:
>>
>>
>>
>> > This did get me thinking though. If the community *did* want to score 
>> highly 
>> > on some of these metrics, what would those be?
>>
>> I'll be happy so long as Clojure is the popular choice for doing the 
>> things where it's advantages should matter: machine learning, AI, NLP, 
>> concurrent programming. 
>>
>> It drives me crazy that Python is doing so well in all of the areas where 
>> Clojure should be winning. There are such beautiful libraries for working 
>> with vectors and matrices with Clojure, which should obviously help with 
>> NLP, yet people use Python instead. Likewise, so much of machine learning 
>> should be done as work in parallel, and Clojure makes that easy, yet Python 
>> is preferred. Drives me crazy. 
>>
>> These last few years I've been at a lot of NLP startups, and the choice 
>> of Python makes me sad. 
>>
>>
>>
>>
>> On Wednesday, March 22, 2017 at 7:17:10 PM UTC-4, Luke Burton wrote:
>>>
>>>
>>> On Mar 22, 2017, at 2:26 PM, Gregg Reynolds <d...@mobileink.com> wrote:
>>>
>>> very interesting stuff, esp. the sociological bits:
>>>
>>> http://stackoverflow.com/insights/survey/2017
>>>
>>> sadly, clojure does not even rank in popularity.  but it's number 1 in 
>>> pay worldwide.  o sweet vengeance!
>>>
>>>
>>> Some fun reading in there, Clojure features a couple of times. It would 
>>> be fun to watch for spikes in traffic to Clojure related resources, because 
>>> I'm sure that landing "most highly paid" will cause a few people to sit up 
>>> and take notice.
>>>
>>> This did get me thinking though. If the community *did* want to score 
>>> highly on some of these metrics, what would those be? Or do none of them 
>>> adequately capture what is valued by the Clojure community?
>>>
>>> I think I'd claim that popularity is a terrible metric, even though it 
>>> can be gratifying to be popular. The fact that lots of people do a 
>>> particular thing doesn't mean that thing is inherently good, or worth 
>>> striving for. Some very popular things are bad lifestyle choices, like 
>>> smoking, a diet high in sugary foods, and writing JavaScript.
>>>
>>> Conversely some very, very good things can die from even the perception 
>>> of being unpopular. We often get people asking on the subreddit why they 
>>> find so many "abandoned" libraries in Clojure. The fact a piece of software 
>>> might have been written years ago, and still be perfectly usable, is such 
>>> an anomaly in more "popular" languages that people assume we've all curled 
>>> up and died. I recently had a project steered away from Clojure (suffice to 
>>> say it was a very good fit, I thought) due to concerns around the 
>>> availability of Clojure programmers in the long term. In Silicon Valley. 
>>> Where you can throw a rock in the air and be certain it will hit a 
>>> programmer on the way down.
>>>
>>> Anyway, my personal metric for Clojure success would be: "for projects 
>>> where Clojure is an appropriate technical fit, how often are you able to 
>>> choose Clojure?" It's a selfish metric but the higher it goes, the happier 
>>> I am ;)
>>>
>>> Luke.
>>>
>>

-- 
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group, send email to clojure@googlegroups.com
Note that posts from new members are moderated - please be patient with your 
first post.
To unsubscribe from this group, send email to
clojure+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/clojure?hl=en
--- 
You received this message because you are subscribed to the Google Groups 
"Clojure" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to clojure+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to