Obligatory XKCD: https://xkcd.com/303/
The number is cool.
Everybody needs a 303.
Ask HN: What's the largest amount of bad code you have ever seen work?
401–410 of 601 posts
Re: Ask HN: What's the largest amount of bad code you have ever seen work?
#402The worst for me was a rescue project: a site for a US tech sector Public-private partnership. Nothing too complicated: recurring donations, paid events, a small members area. They had sent it to an Indian firm to build in Drupal 7 - not a lightweight system to begin with.
I would like to say "cue the stereotypes for Indian developers" and we could all have a good laugh. But no. This is more like Heart of Darkness. They must have traveled to the darkest corners of the subcontinent to find a mind capable of the eldritch horrors we found there. We started keeping a wiki of design patterns, to save us WTF time. Here are a few.
* Memory management. What's that? This site required 16 GIGABYTES as a PHP memory limit in order to load the front page for an anonymous user.
* Security. What's that? Part of the reason it required so much memory, is that it would include a complete debug log of the site's entire transaction history, including PII and credit card numbers, in every session. Meaning any vulnerability was a critical vulnerability.
* They would arbitrarily include the entirety of Zend framework to access a single simple function. This happened several places in the codebase, with different versions of Zend committed.
* Can't reach the ERP to get the price for an event? Let's set it to $999,999.00 and proceed with the transaction.
* Invoice numbers were random numbers between 1-1000. Clashing numbers meant a fatal exception that would fail to store the invoice or payment record... But not until after payment had been processed. Birthday paradox means this happened a lot.
* The developers used arcane bits of Drupal API to do totally mundane things. Like, if you know about hook_page_alter, you know there's a setting in the UI for the frontpage. But we'll just use hook_page_alter instead.
* Write queries using Drupal Views, rewrite the query in code, override the views query with their custom (identical) version using an unusual API hook, just to add a sort.
I could go on, but I think you get the picture. Eldritch horror of a codebase.
Re: Ask HN: What's the largest amount of bad code you have ever seen work?
#403Earlier quoted context omitted.
> Tests that run for 30 hours is an indication that nobody bothered writing unittests. Yes, they were not unit tests. There was no culture of unit tests in the Oracle Database development team. A few people called it "unit tests" but they either said it loosely or they were mistaken. Unit test would not have been effective because every area of the code was deeply entangled with everything else. They did have the con…
I'm amazed that it had that many tests that took that long, but ONLY had 80-95% coverage. I understand the product is huge, but that's crazy.
I do. It's about state coverage: every Boolean flag doubles the possible state of that bit of code: now you need to run everything twice to retain the coverage.
FWIW, I know people who work on SQL processing (big data Hive/Spark, not RDMBS), and a recurrent issue is that an optimisation which benefits most people turns out to be pathologically bad for some queries for some users. Usually those with multiple tables with 8192 columns and some join which takes 4h at the best of times, now takes 6h and so the overnight reports aren't ready in time. And locks held in the process are now blocking some other app which really matters to the businesses existence. These are trouble because they still "work" in the pure 'same outputs as before', it's just the side effects can be be so disruptive.
Re: Ask HN: What's the largest amount of bad code you have ever seen work?
#404Earlier quoted context omitted.
That is absolutely insane. I can't even begin to imagine the complexity of that codebase. I thought my Rails test suite was slow because it takes 4 minutes. If I wrote it in C or C++ it would probably be 10 seconds. I can't imagine a C/C++ application where the test suite takes 20-30 hours on a test farm with 100-200 servers. And if you can break 100-1000 tests with a single change, it doesn't like things are very mo…
Tests in C/C++ run shockingly fast. I ported an application from Ruby to C++ and the tests ran in well under a second when it was taking 10+ seconds in Ruby. Granted because of C++'s type system there were fewer tests, but it was fast enough that I kept thinking something was wrong.
It's because Ruby is only of the slowest languages out there, and C/C++ is usually #1-#2 on many benchmarks.
Re: Ask HN: What's the largest amount of bad code you have ever seen work?
#405Earlier quoted context omitted.
(terminology nazi mode) "... is an indication that the software design isn't (very) decoupled ". You can be modular without being properly decoupled from the other modules.
But then are your modules really modular?
In a C/C++ world, a module is usually defined as a file/.dll/.so on disk. So highly-coupled modules are still modules.
Re: Ask HN: What's the largest amount of bad code you have ever seen work?
#406Earlier quoted context omitted.
Were you even given substantial time to ask the interviewers questions? In most interviews I’ve done, even later round interviews whether it’s a finance company, start-up, FAANG, and companies of all sorts in between, I was given at most 5 minutes to ask questions after some dumb shit whiteboard algo trivia.
I was given 5 minutes to ask questions after each round of interview. That part was ordinary too. That's what most of the other companies do (FAANG or otherwise).
The real risk is for people who are too young to know what to ask
Re: Ask HN: What's the largest amount of bad code you have ever seen work?
#407Earlier quoted context omitted.
Really surprising considering that Oracle is the standard for serious enterprise databases. Not really surprising when you consider Oracle's other bug ridden offerings (probably not as thoroughly tested). Makes me fear for Oracle 18c.
Totally unsurprising if you've ever worked with Oracle. The layers upon layers of legacy cruft are plainly visible even in simple things like the GUI installers.
The fact that the first version shipped in 1979 has to contribute to this as well.
The field of software engineering has matured a lot since then.
Re: Ask HN: What's the largest amount of bad code you have ever seen work?
#408Oracle Database 12.2. It is close to 25 million lines of C code. What an unimaginable horror! You can't change a single line of code in the product without breaking 1000s of existing tests. Generations of programmers have worked on that code under difficult deadlines and filled the code with all kinds of crap. Very complex pieces of logic, memory management, context switching, etc. are all held together with thousand…
I am a current Oracle employee and blame a lot of the mistakes on the overseas development team in india. They are (not all but enough to matter) terrible programmers, but hey when you can throw 10 Indian programmers at a problem for the cost of one American... You can blame your blated mismanaged code base on their management over there. This is likely do to the attrition and generally less talented and less autonom…
I have not found this to be true at all. I have seen both US and Indian developers adding good code as well as ugly code to the Oracle Database product.
The actual damage was done much before I had joined Oracle. It appears that somewhere in the early 2000s, the Oracle codebase went from manageable to sphagetti monster. The changelog showed more changes from US developers than Indian developers at that time. Once the damage was done, all developers whether from the US or India now need to follow this painful process to fix bugs and add features.
Re: Ask HN: What's the largest amount of bad code you have ever seen work?
#409(I am a regular of HN. I don't want this attributed to my acct.) Previous job. Was sysad. This code runs most of the academic US internet. Everything was Perl5. There were over 150 different major applications written with it, ranging from 40 lines to 500k lines. The older the recent commit, the worse. Touching any of this would cause errors, either in itself OR in associated applications! You'd be working on thing A…
You know what's terrible? This could almost be any one of about three of the Perl contracts I've had in the last decade (apart from the US part, obvs.)
Yeah, the more I dig around how perl shops work, this seems endemic.
Now, I work in a heterogenous Windows/Linux shop with non-paleo hardware. We're even deploying on AWS in a limited fashion.
This place has its warts too, but everywhere has something. But that previous place... I'm surprised it still hasn't crashed and burned. Their networking was solid tho. Just anything with Linux was a tire-fire.
And as one more gem, there was a program that terrified the fuck out of me. A fellow engineer showed me this update tool that would update a router remotely. Great. Well, if you add the -n (customer) flag with the customer number, it would update all the routers for that customer!
It was spectacular, and terrifying at the same time. I asked them for their testing procedure, and it was a -t (customer-testing). If they forgot the -testing, well.....
Re: Ask HN: What's the largest amount of bad code you have ever seen work?
#410Oracle Database 12.2. It is close to 25 million lines of C code. What an unimaginable horror! You can't change a single line of code in the product without breaking 1000s of existing tests. Generations of programmers have worked on that code under difficult deadlines and filled the code with all kinds of crap. Very complex pieces of logic, memory management, context switching, etc. are all held together with thousand…
A sentiment among members of a former team was that automated tests meant you didn't need to write understandable code - let the tests do the thinking for you. This, and stuff like your story, are why I don't trust people who promote test-driven development as the best way to write clean APIs.
Yup, in static vs dynamic conversations, I invariably see someone dismiss the value of compiler enforcement by claiming that you should be writing unit tests to cover these cases anyway. Every time I say a silent prayer that I never end up working with the person I'm talking to haha.