584 reputation
26
bio website juliahayward.com
location St Neots, United Kingdom
age
visits member for 2 years, 4 months
seen 2 days ago

Apr
4
comment Why is “Select * from table” considered bad practice
Breaking everything by adding a column is also a good reason why code should always access columns in a datareader by name not by hard-coded ordinal...
Mar
28
answered Pull Frequency Parameters in MVC architecture
Mar
21
comment Functional problems due to localization
thedailywtf.com has plenty of examples of situations like telephone numbers being stored in BIGINTs and ending up being displayed back in scientific notation. Learn from others' fail... :)
Mar
21
answered Functional problems due to localization
Mar
18
comment How important is it to clean up someone else's code when faced with a tight deadline?
Also, training juniors via code review may have a certain amount of time cost now (which as dj18 says should be factored in to your deadlines and estimates anyway) but it will be repaid in due course many times over as it frees you up to do more original work. If your deadlines are so tight that you never have the chance to do this, it smells rather of a death-spiral...
Mar
5
comment What are the advantages of storing xml in a relational database?
The data may not be processable into a fast to query format (nor might you need to query it). Imagine an XML schema where there are hundreds of optional fields of which maybe a handful are ever populated at once. If you insist on modelling this relationally then you'll either end up with vast tables stuffed full of NULLs or the monstrosity that is EAV.
Feb
25
comment How does one regrade a development team appropriately?
12 levels for 20 or so people sounds far too granular. Let me guess, pay rises are predicated solely on moving from one level to another? If so then this is far too rigid. I would sweep away the distinctions and just have trainee, regular and senior, and pay people according to how much they contribute to the business (both in terms of what they produce now, VB6 included, and how proactive they are in preparing for future development).
Feb
19
awarded  Nice Answer
Feb
19
comment What mistakes in managing software products must be avoided to keep people from hating the vendor?
Add to that list inconsistency and tie-in. Make some of your applications do subtly different things when you'd expect them to be the same. (HTML handling in Outlook and IE, perhaps?) Make some of your applications export stuff that can only be read correctly in other applications of yours, and be heavy with proprietary data formats - bonus points if the formats are undocumented and/or have arcane gotchas.
Feb
19
comment “All software has an expiration date by which time it must be rewritten from scratch.”
I'm not sure what you're arguing, then. You seem to be saying that the only reason for an all-out rewrite is incompetence. My point is, there are plenty of factors which simply cannot be anticipated which can force your hand.
Feb
19
answered “All software has an expiration date by which time it must be rewritten from scratch.”
Feb
14
awarded  Yearling
Feb
14
answered Fixing bugs may lead to delayed feedback
Feb
12
answered How does the product owner decide how successful a Sprint was?
Feb
7
comment Does it make sense to write tests for legacy code when there is no time for a complete refactoring?
Too often the "one big refactoring" that's about to happen and that will cure all ills is a myth, concocted by those who simply want to push things they consider boring (writing tests) into the distant future. And if it ever does become real, they will seriously regret having let it become so big!
Jan
28
comment Is it a deal breaker if the HR person asks programming questions during an interview?
"Or, if he/she did read it, does he/she just assume that every applicant is lying on their CV ?" - Enough do (or have their CV "improved" by recruiters) that it would be negligent to assume all CVs are 100% true. Yours might well be but why should you expect special treatment?
Jan
22
answered Should you write down a list of issues, or fix issues as they arise?
Jan
22
comment How to most effectively debug code?
I've heard "We found this bug ourselves and the client hasn't noticed it yet, so just leave it until they do" too many times as well. And having gone on site visits, often the client has noticed, but hasn't reported it. Sometimes because they think there's no point because it won't be fixed, sometimes because they are already looking at a competitor's replacement, and sometimes (rightly or wrongly) "well, it's all a steaming pile of crap anyway".
Jan
22
comment How to most effectively debug code?
And to add to (1), almost every bug that you see in the code implies that there's a bug - or at least an omission - in the test suite. Fix both at the same time and not only do you prove you've fixed the problem at hand, you're safe against it being reintroduced.
Jan
21
comment 2 closest points between 2 rectangles
Even if they are parallel, it is sufficient to check the 16 vertex-vertex distances and the 16 perpendicular distances from one vertex on rectangle A to the extension of a side on rectangle B and vice versa (discarding any where the perpendicular meets outside the rectangle itself). If they are parallel, this gives you a pair of points, not all possible pairs.