On Modalities and Misadventures

Part of the design center of Fishworks is to develop powerful infrastructure that is also easy to use, with the browser as the vector for that usability. So it's been incredibly vindicating to hear some of the initial reaction to our first product, which uses words like "truly breathtaking" to describe the interface.

But as much as we believe in the browser as system interface, we also recognize that it cannot be the only modality for interacting with the system: there remain too many occasions when one needs the lightweight precision of a command line interface. These occasions may be due to usability concerns (starting a browser can be an unnecessary and unwelcome burden for a simple configuration change), but they may also arise because there is no alternative: one cannot use a browser to troubleshoot a machine that cannot communicate over the network or to automate interaction with the system. For various reasons, I came to be the one working on this problem at Fishworks -- and my experiences (and misadventures) in solving it present several interesting object lessons in software engineering.

Before I get to those, a brief aside about our architecture: as will come to no surprise to anyone who has used our applaince and is familiar with browser-based technologies, our interface is AJAX-based. In developing an AJAX-based application, one needs to select a protocol for client-server communication, and for a variety of reasons, we selected XML-RPC -- a simple XML-based remote procedure call protocol. XML-RPC has ample client support (client libraries are readily available for JavaScript, Perl, Python, etc.), and it was a snap to write a C-based XML-RPC server. This allowed us to cleanly separate our server logic (the "controller" in MVC parlance) from our client (the "view"), and (more importantly) it allowed us to easily develop an automated test suite to test server-side logic. Now, (very) early in our development I had written a simple Perl script to allow for us to manually test the server-side that I called "aksh" -- the appliance kit shell. It provided a simple, captive, command-line interface with features like command history, and it gave developers (that is to say, us) the ability to manually tickle their server-side logic without having to write client-side code.

In part because I had written this primordial shell, the task of writing the command line interface fell to me. And when I first approached the problem it seemed natural to simply extend that little Perl script into something more elaborate. That is, I didn't stop to think if Perl was still the right tool for what was a much larger job. In not stopping to reconsider, I was committed a classic software engineering blunder: the problem had changed (and in particular, it had grown quite a bit larger than I understood it to be), but I was still thinking in terms of my existing (outmoded) solution.

As I progressed through the project -- as my shell surpassed 1,000 lines and made its way towards 10,000 -- I was learning of a painful truth about Perl that many others had discovered before me: that it is an undesigned dung heap of a language entirely inappropriate for software in-the-large. As a coping mechanism, I began to vent my frustrations at the language with comments in the source, like this vitriolic gem around exceptions:


eval {
#
# We need to install our own private __WARN__ handler that
# calls die() to be able to catch any non-numeric exception
# from the below coercion without inducing a tedious error
# message as a side-effect. And has it been said recently that
# Perl is a trash heap of a language? Indeed, it reminds one
# of a reeking metropolis like Lagos or Nairobi: having long
# since outgrown its original design (such as there was ever
# any design to begin with), it is now teeming but crippled --
# sewage and vigilantes run amok. And yet, still the masses
# come. But not because it is Utopia. No, they come only
# because this dystopia is marginally better than the only
# alternatives that they know...
#
local $SIG{'__WARN__'} = sub { die(); };
$val = substr($value, 0, length($value) - 1) + 0.0;
};

(In an attempt to prevent roaming gangs of glue-huffing Perl-coding teenagers from staging raids on my comments section: I don't doubt that there's a better way to do what I was trying to achieve above. But I would counter that there's also a way to live like a king in Lagos or Nairobi -- that doesn't make it them tourist destinations.)

Most disconcertingly, the further I got into the project, the more the language became an impediment -- exactly the wrong trajectory for an evolving software system. And so as I wrote more and more code -- and wrestled more and more with the ill-suited environment -- the feeling haunting me became unmistakable: this is the wrong path. There's no worse feeling for a software engineer: knowing that you have made what is likely the wrong decision, but feeling that you can't revisit that decision because of the time already spent down the wrong path. And so, further down the wrong path you go...

Meanwhile, as I was paying the price of my hasty decison, Eric -- always looking for a way to better test our code -- was experimenting writing a test harness in which he embedded SpiderMonkey and emulated a DOM layer. These experiments were a complete success: Eric found that embedding SpiderMonkey into a C program was a snap, and the end result allowed us to get automated test coverage over client JavaScript code that previously had to be tested by hand.

Given both Eric's results and my increasing frustrations with Perl, an answer was becoming clear: I needed to rewrite the appliance shell as a JavaScript/C hybrid, with the bulk of the logic living in JavaScript and system interaction bits living in C. This would allow our two interface modalities (the shell and the web) to commonalize logic, and it would eradicate a complicated and burdensome language from our product. While this seemed like the right direction, I was wary of making another hasty decision. So I started down the new path by writing a library in C that could translate JavaScript objects into an XML-RPC request (and the response back into JavaScript objects). My thinking here was that if the JavaScript approach turned out to be the wrong approach for the shell, we could still use the library in Eric's new testing harness to allow a wider range of testing. As an aside, this is a software engineering technique that I have learned over the years: when faced with a decision, determine if there are elements that are common to both paths, and implement them first, thereby deferring the decision. In my experience, making the decision after having tackled some of its elements greatly informs the decision -- and because the work done was common, no time (or less time) was lost.

In this case, I had the XML-RPC client library rock-solid after about a week of work. The decision could be deferred no longer: it was time to rewrite Perl functionality in JavaScript -- time that would indeed be wasted if the JavaScript path was a dead-end. So I decided that I would give myself a week. If, after that week, it wasn't working out, at least I would know why, and I would be able to return to the squalor of Perl with fewer doubts.

As it turns out, after that week, it was clear that the JavaScript/C hybrid approach was the much better approach -- Perl's death warrant had been signed. And here we come to another lesson that I learned that merits an aside: in the development of DTrace, one regret was we did not start the development of our test suite earlier. We didn't make that mistake with Fishworks: the first test was created just minutes after the first line of code. With my need to now rewrite the shell, this approach paid an unexpected dividend: because I had written many tests for the old, Perl-based shell, I had a ready-made test suite for my new work. Therefore, contrary to the impressions of some about test-driven development, the presence of tests actually accelerated the development of the new shell tremendously. And of course, once I integrated the new shell, I could say with confidence that it did not contain regressions over the old shell. (Indeed, the only user visible change was that it was faster. Much, much, much faster.)

While it was frustrating to think of the lost time (it ultimately took me six weeks to get the new JavaScript-based shell back to where the old Perl-based shell had been), it was a great relief to know that we had put the right architecture in place. And as often happens when the right software architecture is in place, the further I went down the path of the JavaScript/C hybrid, the more often I had the experience of new, interesting functionality simply falling out. In particular, it became clear that I could easily add a second JavaScript instance to the shell to allow for a scripting environment. This allows users to build full, programmatic flow control into their automation infrastructure without ever having to "screen scrape" output. For example, here's a script to display the used and available space in each share on the appliance:


script
run('shares');
projects = list();

printf('%-40s %-10s %-10s\\n', 'SHARE', 'USED', 'AVAILABLE');

for (i = 0; i < projects.length; i++) {
run('select ' + projects[i]);
shares = list();

for (j = 0; j < shares.length; j++) {
run('select ' + shares[j]);

share = projects[i] + '/' + shares[j];
used = run('get space_data').split(/\\s+/)[3];
avail = run('get space_available').split(/\\s+/)[3];

printf('%-40s %-10s %-10s\\n', share, used, avail);
run('cd ..');
}

run('cd ..');
}

If you saved the above to file named "space.aksh", you could run it this way:


% ssh root@myappliance < space.aksh
Password:
SHARE USED AVAILABLE
admin/accounts 18K 248G
admin/exports 18K 248G
admin/primary 18K 248G
admin/traffic 18K 248G
admin/workflow 18K 248G
aleventhal/hw_eng 18K 248G
bcantrill/analytx 1.00G 248G
bgregg/dashbd 18K 248G
bgregg/filesys01 25.5K 100G
bpijewski/access_ctrl 18K 248G
...

(You can also upload SSH keys to the appliance if you do not wish to be prompted for the password.)

As always, don't take our word for it -- download the appliance and check it out yourself! And if you have the appliance (virtual or otherwise), click on "HELP" and then type "scripting" into the search box to get full documentation on the appliance scripting environment!

Comments:

Just scratching my head: Why did you use "perl" for a shell interface ? Since a while we have ksh93 and libshell in OS/Net which were specificcally added to allow custom "shells" to be implemented on top of it (e.g. in a similar way how "dbx" (debgugger with a pdksh-based command-line frontend), "dtksh" (ksh93 with X11/Xt/Motif/CDE addons), "biosh" (ksh93 with bioinformatics add-ons) are implemented).

Posted by Roland Mainz on November 16, 2008 at 07:19 AM PST #

You need to drink the agile programming kool-aid. You got so close, too. In Agile programming, you should have written your first test \*before\* your first line of code.

Posted by Brian Utterback on November 16, 2008 at 10:00 AM PST #

Roland: I'm using the term "shell" here loosely -- from your perspective, it's much better thought of as a captive command-line interface than as a traditional Unix shell. (If you download the appliance and play around with the command line interface, you'll see that it has essentially nothing in common with ksh93.)

Brian: yes -- and don't forget that someone else should have been sitting in my lap the entire time.

Posted by Bryan Cantrill on November 16, 2008 at 10:59 AM PST #

> you'll see that it has essentially nothing in common with ksh93.

Erm... are you sure ? I had to look twice to see the difference between your example and normal ksh93 syntax (the explicit "run" function and that "for" is written "for( expr1; expr2; expr3)" instead of "for(( expr1; expr2; expr3))" are the only major differences, the other things exist in ksh93, too (e.g. OO-style types+functions/methods, string operators, list/tree objects, structured variables etc.)))

Posted by Roland Mainz on November 16, 2008 at 11:12 AM PST #

You turn warnings on, try to pull the last character off a string and coerce it to a number and then complain about figuring out how to suppress the "non numeric string in addition" warning?

Since Perl 5.6 (March 2000) you can turn warnings off within a block:

$val = do { no warnings; substr($value, 0, -1) + 0.0; }

Perl is certainly not perfect, but yours was a poor example of its flaws. (I stopped reading at "tourist destinations.")

Posted by glue-huffing Perl-coding 34-year-old teenager Robert de Forest on November 16, 2008 at 06:11 PM PST #

That snippet was really meant to be more of an example of my frustration than anything else, but let's pick up the issue to which it's referring: the fundamental problem here is not the warnings -- it is the fact that Perl's exception mechanism is unspeakably broken. In most languages, I would expect an illegal coercion to generate an exception that can then be caught and explicitly handled. But because Perl elected to overload die() as an exception mechanism (and because Perl warns on an illegal coercion instead of die()'ing), the only way to achieve the reasonable result is to have the warning actually die(). The code that I quoted was relying on this, but you wouldn't know it from the snippet that I provided here: the next line was a check against $EVAL_ERROR, propagating the "exception" if it's set. So I'm afraid that the "fix" of simply turning off warnings doesn't work here. And even if it did work (that is, even if we were not seeking to explicitly handle the coercion failure), it would still be an unacceptable solution: turning off warnings in Perl allows for too many silent failure conditions to be safely used in production code.

While I happen to think that Perl's busted exception mechanism is a perfectly good example of its flaws, I'll agree that there are so many such examples that it's difficult to choose -- and that this example may indeed be poor relative to Perl's more glaring deformities...

Posted by Bryan Cantrill on November 17, 2008 at 03:03 AM PST #

Raids of teenagers?

As a guy who started coding in the late jurassic, and having managed programmers for a long, long time, I can tell you that the most common reason for "vitriolic comments" comes from ignorance(!) in young programmers. It's a phase, much like adolescence, unrequited crushes and belief in one's own omnipotence and perfection.

Don't get me wrong - such comments usually are way more useful than you'd think - but unfortunately, they never get sent to the guy(s) who implemented the original language/code/system/paradigm. And it's a shame, because the more obvious a language/system/paradigm is, the more easily it will be adopted.

But that never happens.

Instead, such quotes are written as a psychological coping mechanism, one that allows a programmer to blame their own lack of enlightenment on the mistakes of others.

You'll eventually grow out of it, usually at about the time someone else points out your own coding and design failures for about the 100th time.

Posted by some other guy on November 17, 2008 at 04:50 AM PST #

Some other guy,

As I made clear, the comment was indeed a coping mechanism. But I don't believe that I am blaming my "own lack of enlightenment on the mistakes of others", and I'm not sure how exactly one could draw the inference from my comment. Indeed, it sounds like your comment here may well be its own coping mechanism for a deeper pathology, so I'll leave you to it, and wish you well in conquering your personal demons...

Posted by Bryan Cantrill on November 17, 2008 at 04:59 AM PST #

It sounds as replacing Perl with a language you understand better was the right thing for you to do. I think it's worth pointing out, though, at least for the benefit of passers-by, that in Perl you can turn numeric conversion warnings into exceptions simply by writing:

use warnings FATAL => "numeric";

Posted by Robin Houston on November 23, 2008 at 07:39 PM PST #

Ive heard lots about Python being much more productive than Perl. Did you consider Python at all? And if no, why not?

Posted by Kebabbert on November 27, 2008 at 12:01 AM PST #

I just wanted to say I'm deeply gratified to see code comments that are as heated as mine sometimes are, but also more elegant (ahem, less coarse) than what I put down. Also nice to know that limited skills like mine aren't the only just cause of an antipathy for Perl. Thanks!

Posted by Michael Ernest on December 28, 2008 at 02:01 AM PST #

From my perspective it looks like you have been hurt/frustrated by perl and have longed for some reason to leave it. I've met many other programmers with similar stories with various languages (I personally left Java long ago and feel good riddance). But the fact is that millions of other programmers use those very tools successfully and happily. It doesn't mean that they are all stupid for not seeing the same problems that we see. It just means that we're not comfortable using those languages any more and personally feel more comfortable with something else.

The languages and tools we use definitely have pros and cons, but those pros and cons have varying degrees of weight with us due to our own biases and personal experiences. You found a solution that you like and feel comfortable with. If someone else was in your place perhaps they would have stuck with perl (or gone with python or ruby or lisp or ...). My point is that programmers have opinions about what is "right" or "wrong" with a given tool or language, and often those opinions are strong and heated. But unless we are aware of our own biases and admit them, we'll never get passed our own opinions. There is always more than one way to do what we need no matter the tool or language or skill set; there is always a different path. Just because we prefer one method to get to a destination doesn't mean the others are wrong for other people to take.

Now, getting passed that issue- I'm impressed with this project and your solution. I'm a big fan of SpiderMonkey and glad to see yet another use for it. Nice work.

Posted by Spencer Christensen on January 02, 2009 at 03:26 AM PST #

Post a Comment:
Comments are closed for this entry.
About

bmc

Search

Top Tags
Categories
Archives
« April 2014
SunMonTueWedThuFriSat
  
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
   
       
Today