What’s wrong with “Race condition” in India?

[Mustafa Ismail Mustafa]

Thank you, you've hit that nail hard on the head. Its becoming increasingly difficult finding graduates and some people with a year or two of experience under their belt that actually know how to code properly.

"Every time Lotus Notes starts up, somewhere a puppy, a kitten, a lamb, and a baby seal are killed. Lotus Notes is a conspiracy by the forces of Satan to drive us over the brink into madness. The CRC-32 for each file in the installation includes the numbers 666." Gary Wheeler "Programming today is a race between software engineers striving to build bigger and better idiot-proof programs, and the Universe trying to produce bigger and better idiots. So far, the Universe is winning." - Rick Coo

[Alterlife]

To answer your question as it is written: we _do_ use the same term in India. The reason why your interview candidates are not able to respond is because this term isn\'t taught in college. Personally, I think it not important to know this term in order to be a good programmer (especially not with the chutku work that gets shipped to India). This term takes less than a minute to understand, and any programmer who has written a simple multi threaded application has thought through a related problem at some point. My 0.02 INR: I think your intention in asking this question is to figure out if your candidate is familiar with (or can anticipate) problems that can creep into a 'complicated' app. If your candidate does not understand 'race condition', Explain the concept and ask your candidate to respond with an example where this would be a problem in a program.

[Member 4123649]

No we use the same term here also. A race condition occurs when multiple processes access and manipulate the same data concurrently, and the outcome of the execution depends on the particular order in which the access takes place. A race condition is of interest to a hacker when the race condition can be utilized to gain privileged system access.

[Vasudevan Deepak Kumar]

Marc Clifton wrote:

"what is SEX?" My answer would be "Security Enabled XML".

Simply Superb; Innovative Novel and Hilarious. :) :)

Vasudevan Deepak Kumar Personal Homepage
Tech Gossips
A pessimist sees only the dark side of the clouds, and mopes; a philosopher sees both sides, and shrugs; an optimist doesn't see the clouds at all - he's walking on them. --Leonard Louis Levinson

[Vasudevan Deepak Kumar]

Rama Krishna Vavilala wrote:

e

It is just one e which makes the difference between the two islands of information. Just one e to strike the difference in the e-Age.

Vasudevan Deepak Kumar Personal Homepage
Tech Gossips
A pessimist sees only the dark side of the clouds, and mopes; a philosopher sees both sides, and shrugs; an optimist doesn't see the clouds at all - he's walking on them. --Leonard Louis Levinson

[peterchen]

Yep. If someone would stare at me blankly after that, I'd aks "You did multi-threaded programming before?" and if affirmative, "What new problems arise with multi-threaded programming?"

We are a big screwed up dysfunctional psychotic happy family - some more screwed up, others more happy, but everybody's psychotic joint venture definition of CP
blog: TDD - the Aha! | Linkify!| FoldWithUs! | sighist

[peterchen]

I've seen "race condition" used commonly and frequently for the software equivalent. And asking "What problem is reading/writing of the same data by two different threads?" doesn't make a good interview question.

We are a big screwed up dysfunctional psychotic happy family - some more screwed up, others more happy, but everybody's psychotic joint venture definition of CP
blog: TDD - the Aha! | Linkify!| FoldWithUs! | sighist

[Vasudevan Deepak Kumar]

He was meaning the Apartheid, perhaps.

Vasudevan Deepak Kumar Personal Homepage
Tech Gossips
A pessimist sees only the dark side of the clouds, and mopes; a philosopher sees both sides, and shrugs; an optimist doesn't see the clouds at all - he's walking on them. --Leonard Louis Levinson

[RogelioP EX DE HL]

> ask them "what is SEX? Well, be careful what you ask for, someone might have a long enough... TENURE... to know that SEX is a valid mnemonic for a Sign EXtended operation, an official instruction in the Motorola 6809 uP repertoire. Other platforms tried to push for 'SEX' on their instructions lists but efforts were stopped before the stuff got out of the lab.

[Dwayne J Baldwin]

With multiple threads a race condition can occur when two threads are attempting to access the same data. Which one "wins" the race is extremely difficult to test or debug. In .net for example, the last iteration of a worker starts a race between exiting the worker and being cancelled by the owner thread. If the worker wins before setting cancel the owner will not know the thread was cancelled. // bad code if BackgroundWorker.CancellationPending DoWorkEventArgs.cancel // reason for quitting worker endif // race start ...do more work/recursive // race end return result // owner never informed of cancel (race condition) //good code if BackgroudWorker.CancellationPending DoWorkEventARgs.Cancel // either cancel or do more work else ...do more work/recursive // ensures final check with no race condition) end if return result // either cancel handled or thread exited normally

Dwayne J. Baldwin

[Russell Jones]

Do you allow them URGENT Access to code project?

[Ravi Bhavnani]

Imho, it's a basic CS question, regardless of whether the candidate's experience includes writing multithreaded code. However, I agree that getting into the down and dirty nitty gritty details of multithreading would be a waste of time if the position they're trying to fill has nothing to do with writing multithreaded code. /ravi

My new year resolution: 2048 x 1536 Home | Music | Articles | Freeware ravib(at)ravib(dot)com

[realJSOP]

Nascar has race conditions three times every weekend.

"Why don't you tie a kerosene-soaked rag around your ankles so the ants won't climb up and eat your candy ass..." - Dale Earnhardt, 1997
-----
"...the staggering layers of obscenity in your statement make it a work of art on so many levels." - Jason Jystad, 10/26/2001

[Mike Dimmick]

Well, I hope I do, but my skills haven't been seriously tested yet. Most of my MT work has been on single-core systems (generally mobile devices) where there are fewer opportunities to screw up. I'm responsible for a DOS-(style-)emulating library for Windows CE, for porting applications written for DOS-based handhelds, which uses Windows GUI calls to run a window. We would have considered the actual CE console, but Pocket PC devices don't include that library and as far as I know, it's not redistributable. Even if it was available, you don't get control over the menu bars - we don't want the app to be closeable. My first version ran the application code and the UI code on the same thread. Whenever a keypress was requested or a sleep (equivalent), a message pump would be run to process the window's message queue and thereby update the screen. However, this didn't work well with processing-intensive parts of the applications, where you'd have to explicitly slip in a short sleep or direct message pump call to ensure the screen was updated. Also, CE would decide the application wasn't responding (which, of course, it wasn't). So after that had been in the market for a while and stabilised, I moved the UI to a separate thread. That thread creates the window, but Windows' UI processing model requires that keyboard input goes to the active window, so now keyboard input has to move from the UI thread to the application thread. There's some interesting synchronization in there to manage the keyboard buffer. The application thread also writes its text directly into the "screen buffer" structure, then calls InvalidateWindow to tell the window to update itself - I found the overhead of cross-thread window messages too great. There are a few messages used to configure aspects which require full synchronization, for example resizing the screen buffer. In addition the devices have built-in barcode scanners and we have a separate thread which manages that - the manufacturer includes an API which sends scanned data to the window but this was found to be unreliable. Some of our applications respond differently to scanned input compared to the same input entered by hand, so using a keyboard wedge is not appropriate. At some point we have to do something about the single-threaded nature of our application server (the core process is a VB6 standalone EXE, it farms out requests to a pool of worker processes which are themselves single-threaded VB6 ActiveX EXEs). It's just not scaling at present - that is, it's not showing

[Vivek Rajan]

Most Indian developers are not necessarily computer engineers or who otherwise have exposure to computers beyond chat and orkut. In larger companies, all graduates are put into a 3-6 month training program. They are then placed into existing teams writing business logic code in well controlled frameworks. After some time, these developers latch on to a particular vertical (eg retail) and build their career around implementing more business rules in even more well controlled and abstracted frameworks. Of course, not all jobs are like this - just the overwhelming majority. If you want to ask questions on race conditions, you need to interview students from computer science background and from top universities.

[hhexo]

There is a variety of ways they could have heard the term, even in different contexts. For example, the first thing I thought about when reading your question was "Ah, yes, it's the problem with Verilog". That's because I work on an EDA tool that parses hardware descriptions written in the Verilog language, which is prone to race conditions. Then I thought about multithreading. So I find strange that every candidate hasn't heard it somehow. I believe however that one should not draw the conclusion that "if they haven't heard about this technical term, they don't know what multithreading is". My impression is that software engineering is still a bit chaotic when it comes to naming things. In a text you may find the term "race condition" actually defined, in another one you may find that it is classified as just another problem arising from the pre-emptiveness of the thread system.

-+ HHexo +-

[_Zorro_]

Nishant Sivakumar wrote:

My best guess is that the candidates did not get your accent. Sometimes Indians find it hard to understand a foreign accent, specially if the foreigner himself is not himself or herself a native English speaker.

This is weird, I got to work with a lot of Indians, not in India altough. And they got the worse accents I ever heard. They didn't do any effort to speak better, I just had to do my best effort to try to understand what they where saying, so the least I expect, is the same in the other way. Is it like this in India too? Do they have this terrible accent?

[Vasudevan Deepak Kumar]

Perhaps a suggestion for http://www.questpond.com/[^]

Vasudevan Deepak Kumar Personal Homepage
Tech Gossips
A pessimist sees only the dark side of the clouds, and mopes; a philosopher sees both sides, and shrugs; an optimist doesn't see the clouds at all - he's walking on them. --Leonard Louis Levinson

[Lost User]

Marc Clifton wrote:

The equivalent of a race condition in software is not a deadlock but rather the reading/writing of the same data by two different threads. Locking is an attempt to prevent race conditions, and deadlocks are the result of improper locking.

Yep.

Marc Clifton wrote:

So, unless someone has some hardware experience, I'm not surprised that they wouldn't know the term "race condition".

If sonmeones done any driver work they wil be very familiar with this too.

Morality is indistinguishable from social proscription

[Vasudevan Deepak Kumar]

peterchen wrote:

but it works on m computer

Truly. I have encountered such people coming for the interviews very many times. Also, they give replies based on their own 'internal framework'.

Vasudevan Deepak Kumar Personal Homepage
Tech Gossips
A pessimist sees only the dark side of the clouds, and mopes; a philosopher sees both sides, and shrugs; an optimist doesn't see the clouds at all - he's walking on them. --Leonard Louis Levinson