Open Side Menu Go to the Top
Register
The POG Programmers Containment Thread The POG Programmers Containment Thread

09-25-2010 , 07:54 PM
We have like 30 QA people I think. I would not like to go back to doing development without QA. There is a certain skill to creating software breaking.

Also **** MFC. Writing C++ code for windows makes me want to stab things
The POG Programmers Containment Thread Quote
09-25-2010 , 08:03 PM
Quote:
Originally Posted by well named
Also **** MFC. Writing C++ code for windows makes me want to stab things
Sure. This client pays quite promptly though.
The POG Programmers Containment Thread Quote
09-25-2010 , 09:31 PM
Quote:
Originally Posted by Chips Ahoy
Sure. This client pays quite promptly though.
What? Clients pay? wtf is that?

I need QA, because I'm epically bad at testing my work. I'm like all the bad programmer testing cliches jammed in to one body.

I got my first programming job just over 7 years ago, and within 3 months, my boss was telling me I needed to do a better job testing my work. Nothing at all has changed. I still can't consistently test the cases that aren't the most direct path to getting something done. I have to really focus to come up with ways people might do things that aren't right.
The POG Programmers Containment Thread Quote
09-25-2010 , 09:36 PM
Quote:
Originally Posted by well named
We have like 30 QA people I think. I would not like to go back to doing development without QA. There is a certain skill to creating software breaking.

Also **** MFC. Writing C++ code for windows makes me want to stab things
Meh I strongly belive in test first or test often programming and regular functional and unit tests coupled with pair programming of sorts (rotating pairs where one guy has to read through all commits by the other guy and ok them before they get commited work pretty well, it's also great at making all people better programmers which is +EV as well) and regular bug hunting parties. And obviously the good old elitist "good programmers = less testing"
-> QA is really only for the final touches and to add some extra fresh eyes just in case

The test case archive is probably one of the most valuable but underrated items for some software companies (that work on large projects from some huge stock of code that is customized)
The POG Programmers Containment Thread Quote
09-25-2010 , 09:43 PM
we have automated unit test frameworks and automated QA regression tests but probably the bottom line is you can't actually find 100 programmers good enough that you don't need QA, elitist or not.
The POG Programmers Containment Thread Quote
09-25-2010 , 10:17 PM
Quote:
Originally Posted by Zurvan
What? Clients pay? wtf is that?
Invoice quite a bit more than they expected, paid within a week. Some run good.

Quote:
I need QA, because I'm epically bad at testing my work. I'm like all the bad programmer testing cliches jammed in to one body.
Good QA is obviously great -- having to fix whatever problems you find is great incentive to not look too hard. I've seen QA that is worse than nothing though.

I don't have the luxury of QA now.
The POG Programmers Containment Thread Quote
09-25-2010 , 11:52 PM
Automated testing is great for "does it do what it's supposed to". You need actual people to find "What happens when you do what you're not supposed to"

Also, automated testing is impossible when you have no clear requirements, but that's a whole other bitch
The POG Programmers Containment Thread Quote
09-26-2010 , 12:17 AM
and after the people find out what happens when they do what they weren't supposed to and we fix it then we add an automated regression test
The POG Programmers Containment Thread Quote
09-26-2010 , 12:43 AM
Quote:
Originally Posted by Zurvan
What? Clients pay? wtf is that?

I need QA, because I'm epically bad at testing my work. I'm like all the bad programmer testing cliches jammed in to one body.

I got my first programming job just over 7 years ago, and within 3 months, my boss was telling me I needed to do a better job testing my work. Nothing at all has changed. I still can't consistently test the cases that aren't the most direct path to getting something done. I have to really focus to come up with ways people might do things that aren't right.
These days my biggest client takes a month to pay, but a new one I'm picking up is really good at cutting checks ASAP. That will be nice.

Though one little group I did a job for had to wait to raise the money to pay me back. Ah, non-profits.
The POG Programmers Containment Thread Quote
09-26-2010 , 06:30 AM
Quote:
Originally Posted by clowntable
Meh I strongly belive in test first or test often programming and regular functional and unit tests coupled with pair programming of sorts (rotating pairs where one guy has to read through all commits by the other guy and ok them before they get commited work pretty well, it's also great at making all people better programmers which is +EV as well) and regular bug hunting parties. And obviously the good old elitist "good programmers = less testing"
-> QA is really only for the final touches and to add some extra fresh eyes just in case

The test case archive is probably one of the most valuable but underrated items for some software companies (that work on large projects from some huge stock of code that is customized)
good programmers would require less testing imo.
The POG Programmers Containment Thread Quote
09-27-2010 , 09:08 AM
Not testing thoroughly drives me crazy. We have one programmer that just can't grasp it. He'll make it do exactly what it needs to do and that's it. Sometimes he doesn't even do that. He'll just put some code in and tell me it's done without even punching a button. I can break his stuff without even trying hard. This has been going on for 12 years. It's maddening.
The POG Programmers Containment Thread Quote
09-27-2010 , 12:28 PM
well named,

Have you done anything with SQLite? I'm looking for a way around not having foreign key support.
The POG Programmers Containment Thread Quote
09-27-2010 , 01:25 PM
I think the standard workaround for not having foreign keys is the "do it in the code" methodology. I don't think it's usually much of an issue unless there is a lot of concurrent updates/deletes.

I haven't used SQLite but mysql with MyISAM also doesn't have foreign key constraints
The POG Programmers Containment Thread Quote
09-27-2010 , 01:27 PM
Built in foreign key support is for girly men

I have no idea what you're doing, but if there's a framework that will work for your app, then a lot of them handle foreign keys for you.
The POG Programmers Containment Thread Quote
09-27-2010 , 01:32 PM
I AM WRITING A STORED PROCEDURE TO MOCK YOU RIGHT NOW ZURVAN

PREPARE FOR MOCKAGE
The POG Programmers Containment Thread Quote
09-27-2010 , 01:48 PM
The DBA at my old job claimed that stored procedure's were a security risk, and refused to allow them

True story
The POG Programmers Containment Thread Quote
09-27-2010 , 02:15 PM
Quote:
Originally Posted by PartysOver
well named,

Have you done anything with SQLite? I'm looking for a way around not having foreign key support.
You can work around this with triggers
Edit: googled this example:
http://justatheory.com/computers/dat..._triggers.html

the more iimportant thing is...we need to start the SQL debate...
PostgreSQL ALL THE WAY BABY...oh it's not fast enough for you...I call bull**** and/or crappy SQL
Sometimes use SQLite for quick prototyping etc

Oracle is a pure and utter nightmare of "well we do it this way tam, tam, tam". If you need to use it because some customer really wants to go for it better have some dedicated guy working as a one man "Oracle sucks" containment center²

Last edited by clowntable; 09-27-2010 at 02:25 PM. Reason: ²) Make sure you're not that guy
The POG Programmers Containment Thread Quote
09-29-2010 , 09:39 AM
So, I've gone from being up to my neck in alligators to having nothing to do at all in the span of one day. Not sure which one is worse.
The POG Programmers Containment Thread Quote
09-29-2010 , 09:40 AM
Anyone have an idea how the heck he did this?

http://forumserver.twoplustwo.com/62...-posts-885916/
The POG Programmers Containment Thread Quote
09-29-2010 , 09:58 AM
I'm about 99% sure admins can adjust post count in vbulletin

I could go check, but I won't
The POG Programmers Containment Thread Quote
09-29-2010 , 10:57 AM
You could always manually change the post count directly in the database
The POG Programmers Containment Thread Quote
09-30-2010 , 10:42 AM
I need help on my homework

I need to implement IDA* algorithm to solve a fifteen puzzle in C.

I'm really lost. PM me if you can help, even a little bit.
The POG Programmers Containment Thread Quote
09-30-2010 , 10:51 AM
looks like computer science to me. I just writes code

it looks like implementing the algorithm should be simple, I mean the pseudo-code on the page basically gives the idea.

What you have to figure out is what set of heuristics you should use to solve the puzzle, and I guess iterations in this case would be like trying all the possible move combinations to solve the puzzle in N moves, then N+1, or whatever

Caveat: I ain't got no computer science edumacation
The POG Programmers Containment Thread Quote
09-30-2010 , 11:11 AM
I don't got no stinking degree, either

But if you're stuck on something but have pseudocode, then you can try this to get some understanding:

1. Rewrite the pseudocode in plain english, to make sure you understand it.
1a. If that pseudocode is not exactly what you need for your case, edit it as required
2. Put the pseudocode in to your IDE
3. Comment out a line of pseudocode, and replace it with real code
4. Rinse/repeat until the work is done

This is a handy way to solve tough problems, because it forces you to organize all your thoughts in plain english, so you get the understanding of what needs to happen, then you can start thinking about the code one line at a time
The POG Programmers Containment Thread Quote
09-30-2010 , 11:17 AM
I has a VBA question. I have an excel spreadsheet that uses a macro to calculate a bunch of numbers (percentages and dollar amounts) and displays them in a chart. In certain circumstances, I would like to stick a footnote on certain entries. My question is, how can I append the footnote call number to the entry in the chart? Do I have to convert it to a string somehow and then add the call number? This seems unlikely, but is there any way I can display the footnote call number in the chart but still have it be possible to use the number in the cell for calculations?
The POG Programmers Containment Thread Quote

      
m