Language ranking

Language ranking

[Simon Wright]

This[1] ranks 50 languages in terms of "I often get angry when writing
code in this language"; Ada comes 13th.

I particularly like the well-known Irish language O'Caml[2]!

[1]
http://hammerprinciple.com/therighttool/statements/i-often-get-angry-when-writing-code-in-this-langua
[2] http://hammerprinciple.com/therighttool/items/o-caml

Re: Language ranking

[Simon Clubley]

On 2014-04-02, Simon Wright <simon@pushface.org> wrote:
> This[1] ranks 50 languages in terms of "I often get angry when writing
> code in this language"; Ada comes 13th.
>
> I particularly like the well-known Irish language O'Caml[2]!
>
> [1]
> http://hammerprinciple.com/therighttool/statements/i-often-get-angry-when-writing-code-in-this-langua
> [2] http://hammerprinciple.com/therighttool/items/o-caml

I couldn't help but notice that Ada is considered more annoying than C and
C is considered more annoying than Pascal. :-)

I wonder what they mean by "Assembler" ? That can cover a wide range of
architectures from the PIC18 (seriously lousy IMHO) to the elegant ones
such as ARM.

On a more serious note, it would be much more meaningful to have grouped
those languages by typical problem domains because comparing COBOL to
Javascript to Assembler doesn't really tell you anything.

Simon.

-- 
Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
Microsoft: Bringing you 1980s technology to a 21st century world

Re: Language ranking

[Robert A Duff]

Simon Clubley <clubley@remove_me.eisner.decus.org-Earth.UFP> writes:

> On 2014-04-02, Simon Wright <simon@pushface.org> wrote:
>> http://hammerprinciple.com/therighttool/statements/i-often-get-angry-when-writing-code-in-this-langua
>
> I couldn't help but notice that Ada is considered more annoying than C and
> C is considered more annoying than Pascal. :-)

It's not clear what that survey is measuring, but it's not measuring
which languages are considered more annoying.

Consider that I can name a number of annoying things about Ada,
but I can't name a single annoying thing about Agda, which is
the next item on their list.  Is that because Ada is more annoying
than Agda, or because I "consider" it more annoying?"  No, it's
because I don't know anything about Agda!  ;-)

I don't see any data indicating to what extent the people being surveyed
know what they're talking about (though I admit I didn't look very
hard).

I know 32 of those 50 languages well enough to be able to name
something seriously annoying, which leads me to believe that Agda
and the others are annoying, too.

> I wonder what they mean by "Assembler" ? That can cover a wide range of
> architectures from the PIC18 (seriously lousy IMHO) to the elegant ones
> such as ARM.

True.  But pretty much all assembly languages have a lot in common.

> On a more serious note, it would be much more meaningful to have grouped
> those languages by typical problem domains because comparing COBOL to
> Javascript to Assembler doesn't really tell you anything.

Perhaps, but we are in the infancy of programming language design,
and all of those languages have serious flaws that are independent
of application area.  Which application area doesn't benefit
from readability?  Which application area benefits from an
error-prone language?

- Bob

Re: Language ranking

[Shark8]

On 03-Apr-14 08:05, Robert A Duff wrote:
> Which application area benefits from an
> error-prone language?

Operating Systems, general applications?
(Anything that C or C++ is commonly used for in "the industry".)

Re: Language ranking

[Jeffrey Carter]

On 04/03/2014 08:05 AM, Robert A Duff wrote:
>
> Which application area benefits from an
> error-prone language?

Anywhere increased development expense results in increased profits (DOD 
cost-plus contracts) or where making errors and then fixing them results in 
greater profits than not making the errors in the first place (Buy our word 
processor! Now buy an update to fix some of the errors in our word processor!)

-- 
Jeff Carter
"If I could find a sheriff who so offends the citizens of Rock
Ridge that his very appearance would drive them out of town ...
but where would I find such a man? Why am I asking you?"
Blazing Saddles
37

Re: Language ranking

[Randy Brukardt]

"Robert A Duff" <bobduff@shell01.TheWorld.com> wrote in message 
news:wcc38huqwq0.fsf@shell01.TheWorld.com...
...
> Which application area doesn't benefit
> from readability?  Which application area benefits from an
> error-prone language?

Apparently that's the case with web applications. :-)

                   Randy.

Re: Language ranking

[Randy Brukardt]

"Robert A Duff" <bobduff@shell01.TheWorld.com> wrote in message 
news:wcc38huqwq0.fsf@shell01.TheWorld.com...
...
> Perhaps, but we are in the infancy of programming language design,
> and all of those languages have serious flaws that are independent
> of application area.  Which application area doesn't benefit
> from readability?  Which application area benefits from an
> error-prone language?

I bet you didn't expect everyone to semi-seriously answer those questions!

The truth is, history suggests that lousy solutions end up more successful 
than truly well-designed solutions. Must be some sort of ingrained human 
thing. (Makes me sad to be human. :-)

                                   Randy.

Re: Language ranking

[Robert A Duff]

Shark8 <OneWingedShark@gmail.com> writes:

> On 03-Apr-14 08:05, Robert A Duff wrote:
>> Which application area benefits from an
>> error-prone language?
>
> Operating Systems,

I don't see how (for example) implicit conversions that lose
information are useful for writing operating systems.
Or case/switch statements that lack full-coverage checks.
Or any number of other malfeatures that exist in various
programming languages.

>... general applications?

I'm not sure what you mean by that.

> (Anything that C or C++ is commonly used for in "the industry".)

Is that meant to be a definition of "general applications"?

- Bob

Re: Language ranking

[Robert A Duff]

Jeffrey Carter <spam.jrcarter.not@spam.not.acm.org> writes:

> On 04/03/2014 08:05 AM, Robert A Duff wrote:
>>
>> Which application area benefits from an
>> error-prone language?
>
> Anywhere increased development expense results in increased profits (DOD
> cost-plus contracts) or where making errors and then fixing them results
> in greater profits than not making the errors in the first place (Buy
> our word processor! Now buy an update to fix some of the errors in our
> word processor!)

OK, you got me there.  :-(

- Bob

Re: Language ranking

[Robert A Duff]

"Randy Brukardt" <randy@rrsoftware.com> writes:

> I bet you didn't expect everyone to semi-seriously answer those questions!

;-)

> The truth is, history suggests that lousy solutions end up more successful 
> than truly well-designed solutions.

It seems that way.  But not always.

Look to an area that's a bit more mature than computer software.
For example, cars today are a LOT safer than they were in 1964,
and they last longer, too.  I think there's hope for improvement
in the software industry.

>... Must be some sort of ingrained human 
> thing. (Makes me sad to be human. :-)

Better than the alternatives...

- Bob

Re: Language ranking

[Shark8]

On 03-Apr-14 16:12, Robert A Duff wrote:
> Shark8 <OneWingedShark@gmail.com> writes:
>
>> On 03-Apr-14 08:05, Robert A Duff wrote:
>>> Which application area benefits from an
>>> error-prone language?
>>
>> Operating Systems,
>
> I don't see how (for example) implicit conversions that lose
> information are useful for writing operating systems.
> Or case/switch statements that lack full-coverage checks.
> Or any number of other malfeatures that exist in various
> programming languages.

You're absolutely right... I suppose that I should have indicated 
sarcasm above.

>> (Anything that C or C++ is commonly used for in "the industry".)
>
> Is that meant to be a definition of "general applications"?

Basically.
It's also a reference to the general low-emphasis on quality in general 
applications: "we don't have time to do things right" is a too commonly 
heard phrase.

Re: Language ranking

[Jeffrey Carter]

On 04/03/2014 04:21 PM, Robert A Duff wrote:
>
> Look to an area that's a bit more mature than computer software.
> For example, cars today are a LOT safer than they were in 1964,
> and they last longer, too.  I think there's hope for improvement
> in the software industry.

Cars are safer largely because of government regulation. Cars in 1964 weren't 
much safer than they were in 1914.

Will it be 2064 before we get government regulation requiring safer language 
features?

-- 
Jeff Carter
"If I could find a sheriff who so offends the citizens of Rock
Ridge that his very appearance would drive them out of town ...
but where would I find such a man? Why am I asking you?"
Blazing Saddles
37

Re: Language ranking

[Shark8]

On 03-Apr-14 17:43, Jeffrey Carter wrote:
>
> Will it be 2064 before we get government regulation requiring safer
> language features?

That's a seriously bad idea.
Having government regulation would be asking for corruption all along 
that bureaucratic chain.

Re: Language ranking

[tmoran]

> The truth is, history suggests that lousy solutions end up more successful
> than truly well-designed solutions.
    Well-designed takes longer to get out the door than "it worked -
release it to the market."

Re: Language ranking

[tmoran]

> Will it be 2064 before we get government regulation requiring safer language
> features?
    Regulation or product liability - one or the other will start
happening when there's a software bug that painfully hurts a large number
of people.  And it can't be long at this rate before that happens.

Re: Language ranking

[Georg Bauhaus]

On 04/04/14 03:19, Shark8 wrote:
> On 03-Apr-14 17:43, Jeffrey Carter wrote:
>>
>> Will it be 2064 before we get government regulation requiring safer
>> language features?
>
> That's a seriously bad idea.
> Having government regulation would be asking for corruption all along that bureaucratic chain.

*Liability* in software business will suffice.

Regulation, or Law, that would require software to have measurable
properties, and such that would allow customers to sue.

Re: Language ranking

[Robert A Duff]

Shark8 <OneWingedShark@gmail.com> writes:

> You're absolutely right... I suppose that I should have indicated
> sarcasm above.

Ah, I get it now!  ;-)

- Bob

Re: Language ranking

[Shark8]

On 04-Apr-14 01:30, Georg Bauhaus wrote:
> On 04/04/14 03:19, Shark8 wrote:
>> On 03-Apr-14 17:43, Jeffrey Carter wrote:
>>>
>>> Will it be 2064 before we get government regulation requiring safer
>>> language features?
>>
>> That's a seriously bad idea.
>> Having government regulation would be asking for corruption all along
>> that bureaucratic chain.
>
> *Liability* in software business will suffice.
>
> Regulation, or Law, that would require software to have measurable
> properties, and such that would allow customers to sue.

Customers can already sue.
Suing is a civil action; regulation, by definition, is a criminal case 
(in that of a valid regulation, on the part of the accused; in the case 
of illegitimate regulation on the part of the accuser).

Heck, there's already *criminal* law that covers a vast majority of 
these: fraud.

Re: Language ranking

[Randy Brukardt]

<tmoran@acm.org> wrote in message news:lhl9r4$icn$2@speranza.aioe.org...
>> Will it be 2064 before we get government regulation requiring safer 
>> language
>> features?
>    Regulation or product liability - one or the other will start
> happening when there's a software bug that painfully hurts a large number
> of people.  And it can't be long at this rate before that happens.

Hasn't it already happened? Toyota comes to mind. I think they got off easy.

Maybe we need a lot more death - certainly won't be long. (Just look at how 
GM decided to cover up a bad $0.57 part of ten years while people died. 
Especially by redesigning it and not changing the part number so that safety 
regulators couldn't tell that something was different.) But I fear that the 
blame will be placed everywhere but the real culprits ("software is hard" --  
sure, if it's managed badly; "people didn't follow regulations" -- sure, but 
why should people be in the loop at all for software -- that's the job of 
compilers and tools, not people. Etc.)

                             Randy.

Re: Language ranking

[Randy Brukardt]

"Shark8" <OneWingedShark@gmail.com> wrote in message 
news:05m%u.20880$Nj2.13445@fx12.iad...
...
> It's also a reference to the general low-emphasis on quality in general 
> applications: "we don't have time to do things right" is a too commonly 
> heard phrase.

The only solution to that is to hit them in the wallet. "You don't have 
money to do things wrong" would go a long way toward proper quality in 
software. One hopes that the Toyota case would provide a wake-up, but it 
seems that it has not gotten enough coverage. (The general media bought the 
so-called "NASA experts" that weren't even given the full source code to 
analyze. That's just a whitewash. One notes that the juries in the various 
lawsuits didn't buy the whitewash when the plaintiff's lawyers got real 
experts to analyze the real source code. Too bad that didn't get far more 
media play.)

                                             Randy.

Re: Language ranking

[Shark8]

On 04-Apr-14 15:19, Randy Brukardt wrote:
> "Shark8" <OneWingedShark@gmail.com> wrote in message
> news:05m%u.20880$Nj2.13445@fx12.iad...
> ...
>> It's also a reference to the general low-emphasis on quality in general
>> applications: "we don't have time to do things right" is a too commonly
>> heard phrase.
>
> The only solution to that is to hit them in the wallet. "You don't have
> money to do things wrong" would go a long way toward proper quality in
> software. One hopes that the Toyota case would provide a wake-up, but it
> seems that it has not gotten enough coverage.

I used the Toyota case in the executive summary of SQUID as just such an 
illustration; IIRC it was called irrelevant.

> (The general media bought the so-called "NASA experts" that weren't even
> given the full source code to analyze. That's just a whitewash. One notes
> that the juries in the various lawsuits didn't buy the whitewash when the
> plaintiff's lawyers got real experts to analyze the real source code. Too
> bad that didn't get far more media play.)

Fully agreed.

Re: Language ranking

[Yannick Duchêne (Hibou57)]

Le Wed, 02 Apr 2014 09:38:30 +0200, Simon Wright <simon@pushface.org> a  
écrit:

> I particularly like **the well-known Irish language O'Caml**[2]!
>

Are you sure? ;-) OCaml is french, and comes from the INRIA, which stands  
for “Institut National de Recherche en Informatique et Automatisme” and  
translates as “National Institute for Research in Computer Science and  
Automation”.

If you doubt it, have a look at this french site, announcing a new Ocaml  
release (7 months ago) from the INRIA:
http://news.humancoders.com/t/ocaml/items/5657-inria-annonce-ocaml-4-01-0

P.S. If you really enjoy OCaml, you may have a look at SML too.

-- 
“Syntactic sugar causes cancer of the semi-colons.” [1]
“Structured Programming supports the law of the excluded muddle.” [1]
[1]: Epigrams on Programming — Alan J. — P. Yale University

Re: Language ranking

[Adam Beneschan]

On Friday, April 4, 2014 3:37:34 PM UTC-7, Hibou57 (Yannick Duchêne) wrote:
> Le Wed, 02 Apr 2014 09:38:30 +0200, Simon Wright a  
> écrit:

> > I particularly like **the well-known Irish language O'Caml**[2]!

> Are you sure? ;-) OCaml is french,

Simon was joking about the fact that this site spelled it with an apostrophe, making it look like it's an Irish name (like O'Connell).  People do that, sometimes; I'm not really sure why.  I used to work with a person of Japanese descent named Ogata, and some customer reps could never figure out that it wasn't O'Gata.

I did notice they capitalized Ada correctly.  On the other hand, they capitalized "Cobol" and "Javascript" the same way, incorrectly (should be COBOL and JavaScript), so perhaps the fact that they got Ada correct was an accident.

                            -- Adam

Re: Language ranking

[Yannick Duchêne (Hibou57)]

Le Thu, 03 Apr 2014 23:20:37 +0200, Randy Brukardt <randy@rrsoftware.com>  
a écrit:

> "Robert A Duff" <bobduff@shell01.TheWorld.com> wrote in message
> news:wcc38huqwq0.fsf@shell01.TheWorld.com...
> ...
>> Which application area doesn't benefit
>> from readability?  Which application area benefits from an
>> error-prone language?
>
> Apparently that's the case with web applications. :-)
>
>                    Randy.

That's why PHP ranks so well (sigh…)



-- 
“Syntactic sugar causes cancer of the semi-colons.” [1]
“Structured Programming supports the law of the excluded muddle.” [1]
[1]: Epigrams on Programming — Alan J. — P. Yale University

Re: Language ranking

[Yannick Duchêne (Hibou57)]

Le Thu, 03 Apr 2014 19:13:38 +0200, Jeffrey Carter  
<spam.jrcarter.not@spam.not.acm.org> a écrit:

> On 04/03/2014 08:05 AM, Robert A Duff wrote:
>>
>> Which application area benefits from an
>> error-prone language?
>
> Anywhere increased development expense results in increased profits (DOD  
> cost-plus contracts) or where making errors and then fixing them results  
> in greater profits than not making the errors in the first place (Buy  
> our word processor! Now buy an update to fix some of the errors in our  
> word processor!)

I have another theory to explain this, in some way the opposite: anyway  
where “free as in beer” is the most (often the only one) high priority  
criteria and so you need a language which will increase the chance someone  
will make it for free and attract people who believe it's as easily done  
as said. The prerequisites for such a language are well known: write fast,  
no check, no structure, no typing, nothing looking “annoying” and which  
could stop you at any point (no compile time is a bonus too… as  
compilation may fails).


-- 
“Syntactic sugar causes cancer of the semi-colons.” [1]
“Structured Programming supports the law of the excluded muddle.” [1]
[1]: Epigrams on Programming — Alan J. — P. Yale University

Re: Language ranking

[Yannick Duchêne (Hibou57)]

Le Fri, 04 Apr 2014 01:21:58 +0200, Robert A Duff  
<bobduff@shell01.theworld.com> a écrit:

> "Randy Brukardt" <randy@rrsoftware.com> writes:
>
>> I bet you didn't expect everyone to semi-seriously answer those  
>> questions!
>
> ;-)
>
>> The truth is, history suggests that lousy solutions end up more  
>> successful
>> than truly well-designed solutions.
>
> It seems that way.  But not always.
>
> Look to an area that's a bit more mature than computer software.
> For example, cars today are a LOT safer than they were in 1964,
> and they last longer, too.  I think there's hope for improvement
> in the software industry.
>

There is a difference however: most people are OK to give something away  
for a safer car, not for a software. And it's unlikely to change in the  
near future.


-- 
“Syntactic sugar causes cancer of the semi-colons.” [1]
“Structured Programming supports the law of the excluded muddle.” [1]
[1]: Epigrams on Programming — Alan J. — P. Yale University

Re: Language ranking

[Yannick Duchêne (Hibou57)]

Le Thu, 03 Apr 2014 18:26:02 +0200, Shark8 <OneWingedShark@gmail.com> a  
écrit:

> On 03-Apr-14 08:05, Robert A Duff wrote:
>> Which application area benefits from an
>> error-prone language?
>
> Operating Systems, general applications?
> (Anything that C or C++ is commonly used for in "the industry".)

C is special case though, as it's often used as a target language (or a  
portable assembly language). Talking about C, one should always have two  
very distinct cases in mind: hand‑written C and software generated C, and  
not confuse both.


-- 
“Syntactic sugar causes cancer of the semi-colons.” [1]
“Structured Programming supports the law of the excluded muddle.” [1]
[1]: Epigrams on Programming — Alan J. — P. Yale University

Re: Language ranking

[Yannick Duchêne (Hibou57)]

Le Sat, 05 Apr 2014 00:57:24 +0200, Adam Beneschan  
<adambeneschan@gmail.com> a écrit:

> On Friday, April 4, 2014 3:37:34 PM UTC-7, Hibou57 (Yannick Duchêne)  
> wrote:
>> Le Wed, 02 Apr 2014 09:38:30 +0200, Simon Wright a
>> écrit:
>
>> > I particularly like **the well-known Irish language O'Caml**[2]!
>
>> Are you sure? ;-) OCaml is french,
>
> Simon was joking about the fact that this site spelled it with an  
> apostrophe, making it look like it's an Irish name (like O'Connell).

Aaaah, OK… And indeed funny :-) Thanks for the hint Adam.

-- 
“Syntactic sugar causes cancer of the semi-colons.” [1]
“Structured Programming supports the law of the excluded muddle.” [1]
[1]: Epigrams on Programming — Alan J. — P. Yale University

Re: Language ranking

[Shark8]

On 04-Apr-14 17:21, Yannick Duchêne (Hibou57) wrote:
> C is special case though, as it's often used as a target language (or a
> portable assembly language).

There are other languages where you could do this, too.
Forth comes to mind as a possible alternative as it's supposed to be 
pretty dang easy to implement/re-target.

> Talking about C, one should always have two
> very distinct cases in mind: hand‑written C
> and software generated C,
> and not confuse both.

I'm still critical of the SW generated stuff; unless you've done some 
sort of formal verification on the generation itself you're merely 
losing any accuracy/safety in the name of 'portability' because "every 
environment has a C compiler."

Re: Language ranking

[Robert A Duff]

"Yannick Duchêne (Hibou57)" <yannick_duchene@yahoo.fr> writes:

> C is special case though, as it's often used as a target language (or a
> portable assembly language).

I think there is no such thing as a "portable assembly language".
So C isn't one.  ;-)

>...Talking about C, one should always have two
> very distinct cases in mind: hand‑written C and software generated C,
> and  not confuse both.

True.  I think this discussion is about hand-written code.

- Bob

Re: Language ranking

[tmoran]

> > when there's a software bug that painfully hurts a large number of people.
>
> Hasn't it already happened? Toyota comes to mind. I think they got off easy.

How many people do you know who were hurt enough that they "wrote letters
to their Congressman"?  Or insurance companies that charge diffferently
depending on how the software is made.

> blame will be placed everywhere but the real culprits ("software is hard"
  As long as people think "no change in software engineering practices
would help", there will be no change in software engineering practices.

Re: Language ranking

[Jacob Sparre Andersen]

Shark8 <OneWingedShark@gmail.com> wrote:
> On 04-Apr-14 01:30, Georg Bauhaus wrote:

>> *Liability* in software business will suffice.

> Customers can already sue.

Yes.

I don't know about other countries, but in Denmark, you can only sue if
the quality of the product is worse than you should expect.  And
supposedly software can never be perfect[*].

Greetings,

Jacob

[*] Except if it is the accounting software of a major corporation, and
    you complain as a customer of said corporation.
-- 
"If we weren't at least occasionally surprised by the results,
 we might as well save ourselves the trouble of measuring :)"

Re: Language ranking

[Georg Bauhaus]

On 04/04/14 18:42, Shark8 wrote:

>> *Liability* in software business will suffice.
>>
>> Regulation, or Law, that would require software to have measurable
>> properties, and such that would allow customers to sue.
>
> Customers can already sue.

For the most part, no they can't sue. Many software licenses tend
to be long for a reason and at this length they carefully limit
responsibility of licensor for damages the software could cause.

Search them for "LIMITATION", "some states" and "may", or
have a look at how vague an instrument an EULA is.
You can't be a billion dollar company and not be pay a few
lawyers to draft a valid license for a handful of products,
can you? Oh, wait, that's upside down...

Re: Language ranking

[Georg Bauhaus]

On 04/04/14 18:42, Shark8 wrote:

> Heck, there's already *criminal* law that covers a vast majority of these: fraud.


The big corps of today are actively and openly (e.g. Google)
campaigning to undermine what currently is The Law ("users don't
care", they say). They use well known techniques of obfuscating
legal matters, of playing them down.

And they complicate access. E.g., URLs are not linked from legal
text; copy&paste is disabled for text introducing the legally
important matter that is supposed to be establishing a contract;
references that give meaning to context are technically dangling.
Like, "for more on $xyz$, see URL (see blah-blah)".

Practically no end user can handle these supposedly binding
contracts without the assistance of at least one competent lawyer.

And, practically no software billionaire can afford bringing
his software to the required known good state: it might ruin
his business.

If you think about the non-software market, the kind of legal
mess exhibited by ubiquitous software licenses is normally taken
as a sign of misbehavior, if not fraud, since you mentioned the word.

For contrast, a case in question is whether or not a customer
may resell a product he once bought and that he no longer uses...