651 reputation
718
bio website
location
age
visits member for 3 years, 7 months
seen Jul 22 '13 at 7:06

Student at University of Maryland College Park.


Nov
25
comment learning on the clock
and letting programmers learn for some hrs @work might even be cheaper than both paying salary AND the costs of training
Nov
25
comment learning on the clock
I hadnt noticed that interpretation on Hiring for passion, which I see frequently in job posts. If the employer didn t just say it but actually supported it with some time during working hrs, Ill take 'hiring for passion' any day though over the others,
Jan
8
comment cost trade-offs for deploying changes to prod, stored procs vs. LINQ
It seems like you're talking more about worst-case cost than probability? If you're proposing that cost cannot be calculated without risk -- then can risk be calculated without cost?
Jan
8
comment cost trade-offs for deploying changes to prod, stored procs vs. LINQ
@Jarrod, I reworded the title to be consistent with the question as written, which remains to ask about trade-offs, rather than an absolute "which is best".
Jan
7
comment cost trade-offs for deploying changes to prod, stored procs vs. LINQ
+1 but not marked answer since it doesn't answer the question as you say it's wrong; so if you think the question is wrong, then please at least answer the "correct" question.
Jan
7
comment Stored Procedures a bad practice at one of worlds largest IT software consulting firms?
2) on unit testing stored procs. idk about other unit test frameworks but at least with MS Test (VisualStudio.TestTools.UnitTesting), running any of the Assert methods on the stored proc at least requires a Db connection, which by definition makes it more of an integration test than a unit test. And a stored proc may reference state about the database at a global, database level. These may not be fake-able or have interfaces.
Jan
7
comment T-SQL stored procs orders of magnitude faster than LINQ
"runtime performance is irrelevant"? but tell that to the users or developers who spend their time on performance-tuning. Judging by the extended comments yet no answer to the actual question, it seems that either the answer is none or that you don't know?
Jan
7
comment T-SQL stored procs orders of magnitude faster than LINQ
But those JOIN questions (lol) aren't mine. One of the LINQ providers I wrote does support JOINs. stackoverflow.com/a/6350766/266457. And for any functions which don't have a direct T-SQL translation, the Expression tree can still be evaluated as much as possible, and let the remainder of the Expression tree be evaluated in .NET - besides the fact that the query could otherwise be immediately executed, and then run the Regex/financial functions w/ LINQ to objects afterwards.
Jan
7
comment T-SQL stored procs orders of magnitude faster than LINQ
@qes I'm aware of limitations of LINQ, e.g. the following throws an exception in LINQ2SQL VS2010, even though the second part of the BinaryExpression shouldn't need to be evaluated: int? version = null; var books = db.Textbooks.Where(b => version == null || b.Version == (int)version); Trig and financial functions don't seem to be of them. e.g. where is the issue in running the following? var courses = db.Courses.Where(c => Math.Sin(c.Credits) != 1).ToArray(); Most financial (and trig) functions seem to require scalar parameters anyway, not rowsets.
Jan
6
comment T-SQL stored procs orders of magnitude faster than LINQ
@qes: Trig functions not happening? How about financial functions? I definitely see those commonly in stored procs. Iteration w/ state not happening? FETCH. msdn.microsoft.com/en-us/library/ms180152.aspx String manipulation? Last time I checked LINQ to SQL, the String functions were supported. Besides that, string, regex functions that show up in the SELECT statement are easily applied after the call to IQueryProvider.Execute. programmers.stackexchange.com/a/65810/19936
Jan
6
comment T-SQL stored procs orders of magnitude faster than LINQ
@qes: Yes I do. More than 2, but not dozens, and I've written 2. Which are the dozens you've worked with? It seems that simplest table selects cannot be that different across different providers (and besides contradicting "cannot generalize"), how differently can a stored proc represent a simple select than LINQ to SQL can generate the equivalent T-SQL? The LINQ-to-T-SQL translators are already provided by the ORM framework, so where's the fight?
Jan
6
comment T-SQL stored procs orders of magnitude faster than LINQ
@qes: A good comment, but I doubt this. Cannot generalize? Is the generated T-SQL or the algorithms used to generate it from e.g. LINQ-to-SQL and Entity Framework, really so fundamentally different? By time-consuming here I assume you mean development time, not running time (i.e. the question)? Even still, not necessarily. For example, suppose your query involved concatenation, other string manipulation, iteration,with persistence of some state from one row to the next, and calls to trigonometry functions...
Dec
30
comment Not okay to take notes at meetings?
Nice response. From the answers given here, it appears that the issue then is not really with note-taking per se, but the use of a machine which from some may serve as more of a distraction than a tool.
Dec
30
comment Not okay to take notes at meetings?
+1 for sensible justification.
Dec
29
comment Not okay to take notes at meetings?
I like using OneNote.
Dec
29
comment Not okay to take notes at meetings?
Anthony thanks for your input. Before you vote to close this, however, consider other questions which might be applicable to those fields you listed, e.g. programmers.stackexchange.com/questions/24864/…, does that question necessarily need to apply to programmers?
Dec
29
comment Not okay to take notes at meetings?
So do you mean to say that you are against note-taking in general, for these reasons listed? Or that this is a list of some reasons of maybe why. What if points 1-4 are addressed, then are you still against note-taking? Those seem doable.
Dec
16
comment How to deal with people who don't want to share knowledge?
Ask yourself from their perspective, what's the benefit to them and not the team, for putting their time and effort into doing the extra work of documentation?
Dec
3
comment Why don't all companies buy developers the best hardware?
"The difference of productivity between...machines is negligible. The difference in price is significant." What is the value if you were to quantify this negligible difference? -10 min/day ? Suppose the top-end machine is $3000, and assume that almost-top-end is almost-$3000, or $2500. The difference is $500 a fixed cost distributed over 2-3 yrs. Is this your definition of significant?
Dec
3
comment Why don't all companies buy developers the best hardware?
Anandtech has come up with some interesting ways of quantifying performance and cost. anandtech.com/bench/CPU/2