Search Results

Search found 31421 results on 1257 pages for 'software performance'.

Page 284/1257 | < Previous Page | 280 281 282 283 284 285 286 287 288 289 290 291  | Next Page >

  • Passing integer lists in a sql query, best practices

    - by Artiom Chilaru
    I'm currently looking at ways to pass lists of integers in a SQL query, and try to decide which of them is best in which situation, what are the benefots of each, and what are the pitfalls, what should be avoided :) Right now I know of 3 ways that we currently use in our application. 1) Table valued parameter: Create a new Table Valued Parameter in sql server: CREATE TYPE [dbo].[TVP_INT] AS TABLE( [ID] [int] NOT NULL ) Then run the query against it: using (var conn = new SqlConnection(DataContext.GetDefaultConnectionString)) { var comm = conn.CreateCommand(); comm.CommandType = CommandType.Text; comm.CommandText = @" UPDATE DA SET [tsLastImportAttempt] = CURRENT_TIMESTAMP FROM [Account] DA JOIN @values IDs ON DA.ID = IDs.ID"; comm.Parameters.Add(new SqlParameter("values", downloadResults.Select(d => d.ID).ToDataTable()) { TypeName = "TVP_INT" }); conn.Open(); comm.ExecuteScalar(); } The major disadvantages of this method is the fact that Linq doesn't support table valued params (if you create an SP with a TVP param, linq won't be able to run it) :( 2) Convert the list to Binary and use it in Linq! This is a bit better.. Create an SP, and you can run it within linq :) To do this, the SP will have an IMAGE parameter, and we'll be using a user defined function (udf) to convert this to a table.. We currently have implementations of this function written in C++ and in assembly, both have pretty much the same performance :) Basically, each integer is represented by 4 bytes, and passed to the SP. In .NET we have an extension method that convers an IEnumerable to a byte array The extension method: public static Byte[] ToBinary(this IEnumerable intList) { return ToBinaryEnum(intList).ToArray(); } private static IEnumerable<Byte> ToBinaryEnum(IEnumerable<Int32> intList) { IEnumerator<Int32> marker = intList.GetEnumerator(); while (marker.MoveNext()) { Byte[] result = BitConverter.GetBytes(marker.Current); Array.Reverse(result); foreach (byte b in result) yield return b; } } The SP: CREATE PROCEDURE [Accounts-UpdateImportAttempts] @values IMAGE AS BEGIN UPDATE DA SET [tsLastImportAttempt] = CURRENT_TIMESTAMP FROM [Account] DA JOIN dbo.udfIntegerArray(@values, 4) IDs ON DA.ID = IDs.Value4 END And we can use it by running the SP directly, or in any linq query we need using (var db = new DataContext()) { db.Accounts_UpdateImportAttempts(downloadResults.Select(d => d.ID).ToBinary()); // or var accounts = db.Accounts .Where(a => db.udfIntegerArray(downloadResults.Select(d => d.ID).ToBinary(), 4) .Select(i => i.Value4) .Contains(a.ID)); } This method has the benefit of using compiled queries in linq (which will have the same sql definition, and query plan, so will also be cached), and can be used in SPs as well. Both these methods are theoretically unlimited, so you can pass millions of ints at a time :) 3) The simple linq .Contains() It's a more simple approach, and is perfect in simple scenarios. But is of course limited by this. using (var db = new DataContext()) { var accounts = db.Accounts .Where(a => downloadResults.Select(d => d.ID).Contains(a.ID)); } The biggest drawback of this method is that each integer in the downloadResults variable will be passed as a separate int.. In this case, the query is limited by sql (max allowed parameters in a sql query, which is a couple of thousand, if I remember right). So I'd like to ask.. What do you think is the best of these, and what other methods and approaches have I missed?

    Read the article

  • Perf4j Not Logging Correctly

    - by Jehud
    I setup some stop watch calls in my code to measure some code blocks and all the messages are going into my primary log and not into the timing log. The perfStats.log file gets created just fine but all the messages go to the root log which I didn't think was supposed to happen according to the docs I've read. Is there something obvious I'm missing here? perf4j tutorial link: http://perf4j.codehaus.org/devguide.html#Using_the_log4j_Appenders_to_Generate_Real-Time_Performance_Information Example code import org.apache.log4j.Logger; import org.perf4j.LoggingStopWatch; import org.perf4j.StopWatch; public class PerfLogger { /** * @param args */ public static void main(String[] args) { Logger logger = Logger.getLogger(PerfLogger.class.getName()); logger.info("Starting perf log test"); StopWatch stopWatch = new LoggingStopWatch("test time"); try { Thread.sleep(1000); } catch (InterruptedException e) { // TODO Auto-generated catch block e.printStackTrace(); } stopWatch.stop(); } } Example log4j.xml <log4j:configuration xmlns:log4j='http://jakarta.apache.org/log4j/'> <appender name="STDOUT-DEBUG" class="org.apache.log4j.ConsoleAppender"> <layout class="org.apache.log4j.PatternLayout"> <param name="ConversionPattern" value="%d %-5p [%t]%x %M (%F:%L) - %m%n"/> </layout> </appender> <!-- Perf4J appenders --> <!-- This AsyncCoalescingStatisticsAppender groups StopWatch log messages into GroupedTimingStatistics messages which it sends on the file appender defined below --> <appender name="CoalescingStatistics" class="org.perf4j.log4j.AsyncCoalescingStatisticsAppender"> <!-- The TimeSlice option is used to determine the time window for which all received StopWatch logs are aggregated to create a single GroupedTimingStatistics log. Here we set it to 10 seconds, overriding the default of 30000 ms --> <param name="TimeSlice" value="10000"/> <appender-ref ref="fileAppender"/> </appender> <!-- This file appender is used to output aggregated performance statistics --> <appender name="fileAppender" class="org.apache.log4j.FileAppender"> <param name="File" value="perfStats.log"/> <layout class="org.apache.log4j.PatternLayout"> <param name="ConversionPattern" value="%m%n"/> </layout> </appender> <!-- Loggers --> <!-- The Perf4J logger. Note that org.perf4j.TimingLogger is the value of the org.perf4j.StopWatch.DEFAULT_LOGGER_NAME constant. Also, note that additivity is set to false, which is usually what is desired - this means that timing statements will only be sent to this logger and NOT to upstream loggers. --> <logger name="org.perf4j.TimingLogger" additivity="false"> <level value="INFO"/> <appender-ref ref="CoalescingStatistics"/> </logger> <root> <priority value="info"/> <appender-ref ref="STDOUT-DEBUG"/> </root> </log4j:configuration>

    Read the article

  • Is C# slower than VB.NET?

    - by Matt Winckler
    Believe it or not, despite the title, this is not a troll. Running some benchmarks this morning, my colleagues and I have discovered some strange things concerning performance, and I am wondering if we're doing something horribly wrong. We started out comparing C# vs. Delphi Prism calculating prime numbers, and found that Prism was about 30% faster. I figured maybe CodeGear did more optimization when generating IL (the exe was about twice as big as C#'s and had all sorts of different IL in it.) So I decided to write a test in VB.NET as well, assuming that Microsoft's compilers would end up writing essentially the same IL for each language. However, the result there was more shocking: C# was more than three times slower than VB running the same operations. The generated IL was different, but not extremely so, and I'm not good enough at reading it to understand the differences. As a fan of C#, this apparent slowness wounds me horribly, and I am left wondering: what in the world is going on here? Is it time to pack it all in and go write web apps in Ruby? ;-) I've included the code for each below--just copy it into a new VB or C# console app, and run. On my machine, VB finds 348513 primes in about 6.36 seconds. C# finds the same number of primes in 21.76 seconds. (I've got an Intel Core2 Quad Q6600 @2.4Ghz; on another Intel machine in the office the code for both runs much faster but the ratio is about the same; on an AMD machine here the timing is ~10 seconds for VB and ~13 for C#--much less difference, but C# is still always slower.) Both of the console applications were compiled in Release mode, but otherwise no project settings were changed from the defaults generated by Visual Studio 2008. Is it a generally-known fact that C#'s generated IL is worse than VB's? Or is this a strange edge case? Or is my code flawed somehow (most likely)? Any insights are appreciated. VB code Imports System.Diagnostics Module Module1 Private temp As List(Of Int32) Private sw As Stopwatch Private totalSeconds As Double Sub Main() serialCalc() End Sub Private Sub serialCalc() temp = New List(Of Int32)() sw = Stopwatch.StartNew() For i As Int32 = 2 To 5000000 testIfPrimeSerial(i) Next sw.Stop() totalSeconds = sw.Elapsed.TotalSeconds Console.WriteLine(String.Format("{0} seconds elapsed.", totalSeconds)) Console.WriteLine(String.Format("{0} primes found.", temp.Count)) Console.ReadKey() End Sub Private Sub testIfPrimeSerial(ByVal suspectPrime As Int32) For i As Int32 = 2 To Math.Sqrt(suspectPrime) If (suspectPrime Mod i = 0) Then Exit Sub End If Next temp.Add(suspectPrime) End Sub End Module C# Code using System; using System.Collections.Generic; using System.Linq; using System.Text; using System.Diagnostics; namespace FindPrimesCSharp { class Program { List<Int32> temp = new List<Int32>(); Stopwatch sw; double totalSeconds; static void Main(string[] args) { new Program().serialCalc(); } private void serialCalc() { temp = new List<Int32>(); sw = Stopwatch.StartNew(); for (Int32 i = 2; i <= 5000000; i++) { testIfPrimeSerial(i); } sw.Stop(); totalSeconds = sw.Elapsed.TotalSeconds; Console.WriteLine(string.Format("{0} seconds elapsed.", totalSeconds)); Console.WriteLine(string.Format("{0} primes found.", temp.Count)); Console.ReadKey(); } private void testIfPrimeSerial(Int32 suspectPrime) { for (Int32 i = 2; i <= Math.Sqrt(suspectPrime); i++) { if (suspectPrime % i == 0) return; } temp.Add(suspectPrime); } } }

    Read the article

  • Why is processing a sorted array faster than an unsorted array?

    - by GManNickG
    Here is a piece of code that shows some very peculiar performance. For some strange reason, sorting the data miraculously speeds up the code by almost 6x: #include <algorithm> #include <ctime> #include <iostream> int main() { // generate data const unsigned arraySize = 32768; int data[arraySize]; for (unsigned c = 0; c < arraySize; ++c) data[c] = std::rand() % 256; // !!! with this, the next loop runs faster std::sort(data, data + arraySize); // test clock_t start = clock(); long long sum = 0; for (unsigned i = 0; i < 100000; ++i) { // primary loop for (unsigned c = 0; c < arraySize; ++c) { if (data[c] >= 128) sum += data[c]; } } double elapsedTime = static_cast<double>(clock() - start) / CLOCKS_PER_SEC; std::cout << elapsedTime << std::endl; std::cout << "sum = " << sum << std::endl; } Without std::sort(data, data + arraySize);, the code runs in 11.54 seconds. With the sorted data, the code runs in 1.93 seconds. Initially I thought this might be just a language or compiler anomaly. So I tried it Java... import java.util.Arrays; import java.util.Random; public class Main { public static void main(String[] args) { // generate data int arraySize = 32768; int data[] = new int[arraySize]; Random rnd = new Random(0); for (int c = 0; c < arraySize; ++c) data[c] = rnd.nextInt() % 256; // !!! with this, the next loop runs faster Arrays.sort(data); // test long start = System.nanoTime(); long sum = 0; for (int i = 0; i < 100000; ++i) { // primary loop for (int c = 0; c < arraySize; ++c) { if (data[c] >= 128) sum += data[c]; } } System.out.println((System.nanoTime() - start) / 1000000000.0); System.out.println("sum = " + sum); } } with a similar but less extreme result. My first thought was that sorting brings the data into cache, but my next thought was how silly that is because the array was just generated. What is going on? Why is a sorted array faster than an unsorted array? The code is summing up some independent terms, the order should not matter.

    Read the article

  • linux thread synchronization

    - by johnnycrash
    I am new to linux and linux threads. I have spent some time googling to try to understand the differences between all the functions available for thread synchronization. I still have some questions. I have found all of these different types of synchronizations, each with a number of functions for locking, unlocking, testing the lock, etc. gcc atomic operations futexes mutexes spinlocks seqlocks rculocks conditions semaphores My current (but probably flawed) understanding is this: semaphores are process wide, involve the filesystem (virtually I assume), and are probably the slowest. Futexes might be the base locking mechanism used by mutexes, spinlocks, seqlocks, and rculocks. Futexes might be faster than the locking mechanisms that are based on them. Spinlocks dont block and thus avoid context swtiches. However they avoid the context switch at the expense of consuming all the cycles on a CPU until the lock is released (spinning). They should only should be used on multi processor systems for obvious reasons. Never sleep in a spinlock. The seq lock just tells you when you finished your work if a writer changed the data the work was based on. You have to go back and repeat the work in this case. Atomic operations are the fastest synch call, and probably are used in all the above locking mechanisms. You do not want to use atomic operations on all the fields in your shared data. You want to use a lock (mutex, futex, spin, seq, rcu) or a single atomic opertation on a lock flag when you are accessing multiple data fields. My questions go like this: Am I right so far with my assumptions? Does anyone know the cpu cycle cost of the various options? I am adding parallelism to the app so we can get better wall time response at the expense of running fewer app instances per box. Performances is the utmost consideration. I don't want to consume cpu with context switching, spinning, or lots of extra cpu cycles to read and write shared memory. I am absolutely concerned with number of cpu cycles consumed. Which (if any) of the locks prevent interruption of a thread by the scheduler or interrupt...or am I just an idiot and all synchonization mechanisms do this. What kinds of interruption are prevented? Can I block all threads or threads just on the locking thread's CPU? This question stems from my fear of interrupting a thread holding a lock for a very commonly used function. I expect that the scheduler might schedule any number of other workers who will likely run into this function and then block because it was locked. A lot of context switching would be wasted until the thread with the lock gets rescheduled and finishes. I can re-write this function to minimize lock time, but still it is so commonly called I would like to use a lock that prevents interruption...across all processors. I am writing user code...so I get software interrupts, not hardware ones...right? I should stay away from any functions (spin/seq locks) that have the word "irq" in them. Which locks are for writing kernel or driver code and which are meant for user mode? Does anyone think using an atomic operation to have multiple threads move through a linked list is nuts? I am thinking to atomicly change the current item pointer to the next item in the list. If the attempt works, then the thread can safely use the data the current item pointed to before it was moved. Other threads would now be moved along the list. futexes? Any reason to use them instead of mutexes? Is there a better way than using a condition to sleep a thread when there is no work? When using gcc atomic ops, specifically the test_and_set, can I get a performance increase by doing a non atomic test first and then using test_and_set to confirm? *I know this will be case specific, so here is the case. There is a large collection of work items, say thousands. Each work item has a flag that is initialized to 0. When a thread has exclusive access to the work item, the flag will be one. There will be lots of worker threads. Any time a thread is looking for work, they can non atomicly test for 1. If they read a 1, we know for certain that the work is unavailable. If they read a zero, they need to perform the atomic test_and_set to confirm. So if the atomic test_and_set is 500 cpu cycles because it is disabling pipelining, causes cpu's to communicate and L2 caches to flush/fill .... and a simple test is 1 cycle .... then as long as I had a better ratio of 500 to 1 when it came to stumbling upon already completed work items....this would be a win.* I hope to use mutexes or spinlocks to sparilngly protect sections of code that I want only one thread on the SYSTEM (not jsut the CPU) to access at a time. I hope to sparingly use gcc atomic ops to select work and minimize use of mutexes and spinlocks. For instance: a flag in a work item can be checked to see if a thread has worked it (0=no, 1=yes or in progress). A simple test_and_set tells the thread if it has work or needs to move on. I hope to use conditions to wake up threads when there is work. Thanks!

    Read the article

  • Simple XNA 2D demo: why is my F# version slower than C# version?

    - by Den
    When running this XNA application it should display a rotated rectangle that moves from top-left corner to bottom-right corner. It looks like my F# version is noticeably much slower. It seems that the Draw method skips a lot of frames. I am using VS 2012 RC, XNA 4.0, .NET 4.5, F# 3.0. I am trying to make it as functional as possible. What could be the reason for poor performance? C#: class Program { static void Main(string[] args) { using (var game = new FlockGame()) { game.Run(); } } } public class FlockGame : Game { private GraphicsDeviceManager graphics; private DrawingManager drawingManager; private Vector2 position = Vector2.Zero; public FlockGame() { graphics = new GraphicsDeviceManager(this); } protected override void Initialize() { drawingManager = new DrawingManager(graphics.GraphicsDevice); this.IsFixedTimeStep = false; } protected override void Update(GameTime gameTime) { position = new Vector2(position.X + 50.1f * (float)gameTime.ElapsedGameTime.TotalSeconds, position.Y + 50.1f * (float)gameTime.ElapsedGameTime.TotalSeconds); base.Update(gameTime); } protected override void Draw(GameTime gameTime) { //this.GraphicsDevice.Clear(Color.Lavender) drawingManager.DrawRectangle(position, new Vector2(100.0f, 100.0f), 0.7845f, Color.Red); base.Draw(gameTime); } } public class DrawingManager { private GraphicsDevice GraphicsDevice; private Effect Effect; public DrawingManager(GraphicsDevice graphicsDevice) { GraphicsDevice = graphicsDevice; this.Effect = new BasicEffect(this.GraphicsDevice) { VertexColorEnabled = true, Projection = Matrix.CreateOrthographicOffCenter(0.0f, this.GraphicsDevice.Viewport.Width, this.GraphicsDevice.Viewport.Height, 0.0f, 0.0f, 1.0f) }; } private VertexPositionColor[] GetRectangleVertices (Vector2 center, Vector2 size, float radians, Color color) { var halfSize = size/2.0f; var topLeft = -halfSize; var bottomRight = halfSize; var topRight = new Vector2(bottomRight.X, topLeft.Y); var bottomLeft = new Vector2(topLeft.X, bottomRight.Y); topLeft = Vector2.Transform(topLeft, Matrix.CreateRotationZ(radians)) + center; topRight = Vector2.Transform(topRight, Matrix.CreateRotationZ(radians)) + center; bottomRight = Vector2.Transform(bottomRight, Matrix.CreateRotationZ(radians)) + center; bottomLeft = Vector2.Transform(bottomLeft, Matrix.CreateRotationZ(radians)) + center; return new VertexPositionColor[] { new VertexPositionColor(new Vector3(topLeft, 0.0f), color), new VertexPositionColor(new Vector3(topRight, 0.0f), color), new VertexPositionColor(new Vector3(topRight, 0.0f), color), new VertexPositionColor(new Vector3(bottomRight, 0.0f), color), new VertexPositionColor(new Vector3(bottomRight, 0.0f), color), new VertexPositionColor(new Vector3(bottomLeft, 0.0f), color), new VertexPositionColor(new Vector3(bottomLeft, 0.0f), color), new VertexPositionColor(new Vector3(topLeft, 0.0f), color) }; } public void DrawRectangle(Vector2 center, Vector2 size, float radians, Color color) { var vertices = GetRectangleVertices(center, size, radians, color); foreach (var pass in this.Effect.CurrentTechnique.Passes) { pass.Apply(); this.GraphicsDevice.DrawUserPrimitives(PrimitiveType.LineList, vertices, 0, vertices.Length/2); } } } F#: namespace Flocking module FlockingProgram = open System open Flocking [<STAThread>] [<EntryPoint>] let Main _ = use g = new FlockGame() g.Run() 0 //------------------------------------------------------------------------------ namespace Flocking open System open System.Diagnostics open Microsoft.Xna.Framework open Microsoft.Xna.Framework.Graphics open Microsoft.Xna.Framework.Input type public FlockGame() as this = inherit Game() let mutable graphics = new GraphicsDeviceManager(this) let mutable drawingManager = null let mutable position = Vector2.Zero override Game.LoadContent() = drawingManager <- new Rendering.DrawingManager(graphics.GraphicsDevice) this.IsFixedTimeStep <- false override Game.Update gameTime = position <- Vector2(position.X + 50.1f * float32 gameTime.ElapsedGameTime.TotalSeconds, position.Y + 50.1f * float32 gameTime.ElapsedGameTime.TotalSeconds) base.Update gameTime override Game.Draw gameTime = //this.GraphicsDevice.Clear(Color.Lavender) Rendering.DrawRectangle(drawingManager, position, Vector2(100.0f, 100.0f), 0.7845f, Color.Red) base.Draw gameTime //------------------------------------------------------------------------------ namespace Flocking open System open System.Collections.Generic open Microsoft.Xna.Framework open Microsoft.Xna.Framework.Graphics open Microsoft.Xna.Framework.Input module Rendering = [<AllowNullLiteral>] type DrawingManager (graphicsDevice : GraphicsDevice) = member this.GraphicsDevice = graphicsDevice member this.Effect = new BasicEffect(this.GraphicsDevice, VertexColorEnabled = true, Projection = Matrix.CreateOrthographicOffCenter(0.0f, float32 this.GraphicsDevice.Viewport.Width, float32 this.GraphicsDevice.Viewport.Height, 0.0f, 0.0f, 1.0f)) let private GetRectangleVertices (center:Vector2, size:Vector2, radians:float32, color:Color) = let halfSize = size / 2.0f let mutable topLeft = -halfSize let mutable bottomRight = halfSize let mutable topRight = new Vector2(bottomRight.X, topLeft.Y) let mutable bottomLeft = new Vector2(topLeft.X, bottomRight.Y) topLeft <- Vector2.Transform(topLeft, Matrix.CreateRotationZ(radians)) + center topRight <- Vector2.Transform(topRight, Matrix.CreateRotationZ(radians)) + center bottomRight <- Vector2.Transform(bottomRight, Matrix.CreateRotationZ(radians)) + center bottomLeft <- Vector2.Transform(bottomLeft, Matrix.CreateRotationZ(radians)) + center [| new VertexPositionColor(new Vector3(topLeft, 0.0f), color) new VertexPositionColor(new Vector3(topRight, 0.0f), color) new VertexPositionColor(new Vector3(topRight, 0.0f), color) new VertexPositionColor(new Vector3(bottomRight, 0.0f), color) new VertexPositionColor(new Vector3(bottomRight, 0.0f), color) new VertexPositionColor(new Vector3(bottomLeft, 0.0f), color) new VertexPositionColor(new Vector3(bottomLeft, 0.0f), color) new VertexPositionColor(new Vector3(topLeft, 0.0f), color) |] let DrawRectangle (drawingManager:DrawingManager, center:Vector2, size:Vector2, radians:float32, color:Color) = let vertices = GetRectangleVertices(center, size, radians, color) for pass in drawingManager.Effect.CurrentTechnique.Passes do pass.Apply() drawingManager.GraphicsDevice.DrawUserPrimitives(PrimitiveType.LineList, vertices, 0, vertices.Length/2)

    Read the article

  • Craftsmanship is ALL that Matters

    - by Wayne Molina
    Today, I'm going to talk about a touchy subject: the notion of working in a company that doesn't use the prescribed "best practices" in its software development endeavours.  Over the years I have, using a variety of pseudonyms, asked this question on popular programming forums.  Although I always add in some minor variation of the story to avoid suspicion that it's the same person posting, the crux of the tale remains the same: A Programmer’s Tale A junior software developer has just started a new job at an average company, creating average line-of-business applications for internal use (the most typical scenario programmers find themselves in).  This hypothetical newbie has spent a lot of time reading up on the "theory" of software development, devouring books, blogs and screencasts from well-known and respected software developers in the community in order to broaden his knowledge and "do what the pros do".  He begins his new job, eager to apply what he's learned on a real-world project only to discover that his new teammates doesn't use any of those concepts and techniques.  They hack their way through development, or in a best-case scenario use some homebrew, thrown-together semblance of a framework for their applications that follows not one of the best practices suggested by the “elite” in the software community - things like TDD (TDD as a "best practice" is the only subjective part of this post, but it's included here due to a very large following of respected developers who consider it one), the SOLID principles, well-known and venerable tools, even version control in a worst case and truly nightmarish scenario.  Our protagonist is frustrated that he isn't doing things the "proper" way - a way he's spent personal time digesting and learning about and, more importantly, a way that some of the top developers in the industry advocate - and turns to a forum to ask the advice of his peers. Invariably the answer I, in the guise of the concerned newbie, will receive is that A) I don't know anything and should just shut my mouth and sling code the bad way like everybody else on the team, and B) These "best practices" are fade or a joke, and the only thing that matters is shipping software to your customers. I am here today to say that anyone who says this, or anything like it, is not only full of crap but indicative of exactly the type of “developer” that has helped to give our industry a bad name.  Here is why: One Who Knows Nothing, Understands Nothing On one hand, you have the cognoscenti of the .NET development world.  Guys like James Avery, Jeremy Miller, Ayende Rahien and Rob Conery; all well-respected and noted programmers that are pretty much our version of celebrities.  These guys write blogs, books, and post videos outlining the "correct" way of writing software to make sure it not only works but is maintainable and extensible and a joy to work with.  They tout the virtues of the SOLID principles, or of using TDD/BDD, or using a mature ORM like NHibernate, Subsonic or even Entity Framework. On the other hand, you have Joe Everyman, Lead Software Developer at Initrode Corporation - in our hypothetical story Joe is the junior developer's new boss.  Joe's been with Initrode for 10 years, starting as the company’s very first programmer and over the years building up a little fiefdom of his own until at the present he’s in charge of all Initrode’s software development.  Joe writes code the same way he always has, without bothering to learn much, if anything.  He looked at NHibernate once and found it was "too hard", so he uses a primitive implementation of the TableDataGateway pattern as a wrapper around SqlClient.SqlConnection and SqlClient.SqlCommand instead of an actual ORM (or, in a better case scenario, has created his own ORM); the thought of using LINQ or Entity Framework or really anything other than his own hastily homebrew solution has never occurred to him.  He doesn't understand TDD and considers “testing” to be using the .NET debugger to step through code, or simply loading up an app and entering some values to see if it works.  He doesn't really understand SOLID, and he doesn't care to.  He's worked as a programmer for years, and that's all that counts.  Right?  WRONG. Who would you rather trust?  Someone with years of experience and who writes books, creates well-known software and is akin to a celebrity, or someone with no credibility outside their own minute environment who throws around their clout and company seniority as the "proof" of their ability?  Joe Everyman may have years of experience at Initrode as a programmer, and says to do things "his way" but someone like Jeremy Miller or Ayende Rahien have years of experience at companies just like Initrode, THEY know ten times more than Joe Everyman knows or could ever hope to know, and THEY say to do things "this way". Here's another way of thinking about it: If you wanted to get into politics and needed advice on the best way to do it, would you rather listen to the mayor of Hicktown, USA or Barack Obama?  One is a small-time nobody while the other is very well-known and, as such, would probably have much more accurate and beneficial advice. NOTE: The selection of Barack Obama as an example in no way, shape, or form suggests a political affiliation or political bent to this post or blog, and no political innuendo should be mistakenly read from it; the intent was merely to compare a small-time persona with a well-known persona in a non-software field.  Feel free to replace the name "Barack Obama" with any well-known Congressman, Senator or US President of your choice. DIY Considered Harmful I will say right now that the homebrew development environment is the WORST one for an aspiring programmer, because it relies on nothing outside it's own little box - no useful skill outside of the small pond.  If you are forced to use some half-baked, homebrew ORM created by your Director of Software, you are not learning anything valuable you can take with you in the future; now, if you plan to stay at Initrode for 10 years like Joe Everyman, this is fine and dandy.  However if, like most of us, you want to advance your career outside a very narrow space you will do more harm than good by sticking it out in an environment where you, to be frank, know better than everybody else because you are aware of alternative and, in almost most cases, better tools for the job.  A junior developer who understands why the SOLID principles are good to follow, or why TDD is beneficial, or who knows that it's better to use NHibernate/Subsonic/EF/LINQ/well-known ORM versus some in-house one knows better than a senior developer with 20 years experience who doesn't understand any of that, plain and simple.  Anyone who disagrees is either a liar, or someone who, just like Joe Everyman, Lead Developer, relies on seniority and tenure rather than adapting their knowledge as things evolve. In many cases, the Joe Everymans of the world act this way out of fear - they cannot possibly fathom that a “junior” could know more than them; after all, they’ve spent 10 or more years in the same company, doing the same job, cranking out the same shoddy software.  And here comes a newbie who hasn’t spent 10+ years doing the same things, with a fresh and often radical take on the craft, and Joe Everyman is afraid he might have to put some real effort into his career again instead of just pointing to his 10 years of service at Initrode as “proof” that he’s good, or that he might have to learn something new to improve; in most cases the problem is Joe Everyman, and by extension Initrode itself, has a mentality of just being “good enough”, and mediocrity is the rule of the day. A Thorn Bush is No Place for a Phoenix My advice is that if you work on a team where they don't use the best practices that some of the most famous developers in our field say is the "right" way to do things (and have legions of people who agree), and YOU are aware of these practices and can see why they work, then LEAVE the company.  Find a company where they DO care about quality, and craftsmanship, otherwise you will never be happy.  There is no point in "dumbing" yourself down to the level of your co-workers and slinging code without care to craftsmanship.  In 95% of these situations there will be no point in bringing it to the attention of Joe Everyman because he won't listen; he might even get upset that someone is trying to "upstage" him and fire the newbie, and replace someone with loads of untapped potential with a drone that will just nod affirmatively and grind out the tasks assigned without question. Find a company that has people smart enough to listen to the "best and brightest", and be happy.  Do not, I repeat, DO NOT waste away in a job working for ignorant people.  At the end of the day software development IS a craft, and a level of craftsmanship is REQUIRED for any serious professional.  When you have knowledgeable people with the credibility to back it up saying one thing, and small-time people who are, to put it bluntly, nobodies in the field saying and doing something totally different because they can't comprehend it, leave the nobodies to their own devices to fade into obscurity.  Work for a company that uses REAL software engineering techniques and really cares about craftsmanship.  The biggest issue affecting our career, and the reason software development has never been the respected, white-collar career it was meant to be, is because hacks and charlatans can pass themselves off as professional programmers without following a lick of good advice from programmers much better at the craft than they are.  These modern day snake-oil salesmen entrench themselves in companies by hoodwinking non-technical businesspeople and customers with their shoddy wares, end up in senior/lead/executive positions, and push their lack of knowledge on everybody unfortunate enough to work with/for/under them, crushing any dissent or voices of reason and change under their tyrannical heel and leaving behind a trail of dismayed and, often, unemployed junior developers who were made examples of to keep up the facade and avoid the shadow of doubt being cast upon them. To sum this up another way: If you surround yourself with learned people, you will learn.  Surround yourself with ignorant people who can't, as the saying goes, see the forest through the trees, and you'll learn nothing of any real value.  There is more to software development than just writing code, and the end goal should not be just "shipping software", it should be shipping software that is extensible, maintainable, and above all else software whose creation has broadened your knowledge in some capacity, even if a minor one.  An eager newbie who knows theory and thirsts for knowledge can easily be moulded and taught the advanced topics, but the same can't be said of someone who only cares about the finish line.  This industry needs more people espousing the benefits of software craftsmanship and proper software engineering techniques, and less Joe Everymans who are unwilling to adapt or foster new ways of thinking. Conclusion - I Cast “Protection from Fire” I am fairly certain this post will spark some controversy and might even invite the flames.  Please keep in mind these are opinions and nothing more.  A little healthy rant and subsequent flamewar can be good for the soul once in a while.  To paraphrase The Godfather: It helps to get rid of the bad blood.

    Read the article

  • Is there a distributed project management software like Redmine?

    - by Tobias Kienzler
    I am quite familiar with and love using git, among other reasons due to its distributed nature. Now I'd like to set up some similarly distributed (FOSS) Project Management software with features similar to what Redmine offers, such as Issue & time tracking, milestones Gantt charts, calendar git integration, maybe some automatic linking of commits and issues Wiki (preferably with Mathjax support) Forum, news, notifications Multiple Projects However, I am looking for a solution that does not require a permanently accesible server, i.e. like in git, each user should have their own copy which can be easily synchronized with others. However it should be possible to not have a copy of every Project on every machine. Since trac uses multiple instances for multiple projects anyway, I was considering using that, but I neither know how well it adapts to simply giting the database itself (which would be be easiest way to handle the distribution due to git being used anyway), nor does it include all of Redmine's feature. So, can you recommend me a distributed project management software? If your suggestion is a software that usually runs on a server please include a description of the distribution method (e.g. whether simply putting the data in a git repository would do the trick), and if it's e.g. trac, please mention plugins required to include the features mentioned.

    Read the article

  • Is it more valuable to double major in Computer Science/Software Engineering or get an undergraduate CS degree with a Masters in SE?

    - by Austin Hyde
    A friend and I (both in college) are currently in a debate over which is better, in terms of employment opportunities, experience, and education: a Bachelors degree in both Computer Science and Software Engineering, or a Bachelors in Computer Science with a Masters in Software Engineering. My point of view is that I would rather go to school for 4-4.5 years to learn both sides of the field, and be out working on real projects gaining real experience, by going the double major route. His point of view is that it would look better to potential employers if he had a Bachelors in CS and Masters in SE. That way, when he's finally done after 4 years of CS and 2-4 of SE (depending on where he goes), he can pretty much have his choosing of what he wants to do. We are both in agreement on the distinction between the two degrees: CS is "traditional" and about the theory of algorithms, data structures, and programming, where SE is the study of the design of software and the implementation of CS theory. So, what's your stance on this debate? Have you gone one route or another? And most importantly, why?

    Read the article

  • How can I solve display glitches and poor performance with ATI fglrx driver on my ThinkPad X100e?

    - by rewbs
    I noticed that video performance on my Thinkpad X100e was very poor compared to Windows 7, so I installed the ATI fglrx proprietary drivers by using the "Additional Drivers" dialogue box. The system has an ATI Radeon Mobility HD 3200 chip. The result of installing the drivers is pretty devastatingly negative, with symptoms such as skewed content in windows, browser tabs and text boxes failing to refresh when their content changes. In fact, please excuse typos in this post, because I can't really see what I am typing. :) I also notice that HD video playback performance is no better - perhaps even worse - than prior to installing the drivers. Example of what I see: Here's the output of fglrxinfo: display: :0.0 screen: 0 OpenGL vendor string: ATI Technologies Inc. OpenGL renderer string: ATI Radeon HD 3200 Graphics OpenGL version string: 3.3.10237 Compatibility Profile Context Output of lspci | grep -i vga: 01:05.0 VGA compatible controller: ATI Technologies Inc RS780M/RS780MN [Radeon HD 3200 Graphics] I'm on Ubuntu 10.10 with kernel 2.6.35-22-generic-pae. What can I try? Many thanks, -R

    Read the article

  • I'm hoping to start an online supermarket and needs advice on open source shopping cart software and applications..

    - by Betterman Simidi
    I have been researching on both open source software and off-the-shelf software for an online supermarket project in Africa. I have now narrowed by search to X-cart and the PHP based PrestaShop shopping carts. My plan is to acquire an open source shopping cart either by purchasing or as a free open source cart and hire a local developer to customize it to our local needs. I have been doing the demo for x-cart for three weeks now and had thought it might work best for us but after going through the 600 page manual and I'm concerned with how far it can be localized. Yesterday i was looking at the possible Prestashop free open source cart and i seem to like the back-end. Didn't like the back-end for Magento much but from reviews conducted by third-parties they seem to recommend it. I'm now wondering whether i should have a developer start the whole project from scratch, or use an open source software such us PrestaShop or get x-cart which can then be customized. Note that my store will have thousands of products and services including groceries and so i want something that can handle upto 500,000 products and over. Kindly advice. 93276

    Read the article

  • Is there such a thing as "closure" with software work?

    - by Bobby Tables
    I burned out last year (after a decade of fulltime programming jobs) and am on a sabbatical now. With all the self-examination I've started to figure out some of the root causes of my burnout, and one of the major ones is basically this: there was never any real closure in any of the work I've ever done. It was always a case of getting into an open-ended support/maintenance grind and going stale. When I first entered the industry, I had this image of programming work being very project-based. And I expected projects to have a start, beginning, and END. And then you move on and start on something totally new and fresh. Basically I never expected that a lot (most) of software work involves supporting and maintaining the same code base for open-ended long periods of time - years and even decades. That, combined with generally having itchy feet makes me think that burnout is inevitable for me, after 2-3 years, in ANY fulltime software job. All this sounds like I probably should have been a contractor instead of a fulltimer. But when I discuss this with people, a lot of them say that even THEN you can't really escape having to go back and maintain/support the stuff you worked on, over and over (eg. Coming back on support contracts, for example). The nature of software work is simply like that. There is no project closure, unlike in many other engineering fields. So my question is - Is there ANY programming work out there which is based on short to mid term projects/stints and then moving on cleanly? And is there any particular industry domain or specialization where this kind of project work is typical?

    Read the article

  • Are there currently any modern, standardized, aptitude test for software engineering?

    - by Matthew Patrick Cashatt
    Background I am a working software engineer who is in the midst of seeking out a new contract for the next year or so. In my search, I am enduring several absurd technical interviews as indicated by this popular question I asked earlier today. Even if the questions I was being asked weren't almost always absurd, I would be tired nonetheless of answering them many times over for various contract opportunities. So this got me thinking that having a standardized exam that working software professionals could take would provide a common scorecard that could be referenced by interviewers in lieu of absurd technical interview questions (i.e. nerd hazing). Question Is there a standardized software engineering aptitude test (SEAT??) available for working professionals to take? If there isn't a such an exam out there, what questions or topics should be covered? An additional thought Please keep in mind, if suggesting a question or topic, to focus on questions or topics that would be relevant to contemporary development practices and realistic needs in the workforce as that would be the point of a standard aptitude test. In other words, no clown traversal questions.

    Read the article

  • FREE three days of online SharePoint 2010 development training for UK software houses Feb 9th to 11th

    - by Eric Nelson
    I have been working to get a SharePoint development course delivered online in February and March – online means lots of opportunities to ask questions. The first dates are now in place. The training is being delivered as a benefit for companies signed up to Microsoft Platform Ready. It is intended for UK based companies who develop software products* Agenda: Day 1 (Live Meeting 3 hours) 1:30 - 4:30 •         Getting Started with SharePoint: Understand why and how to start developing for SharePoint 2010 •         SharePoint 2010 Developer Roadmap:  Explore the new capabilities and features •         UI Enhancements: How to take advantage of the many UI enhancements including the fluent UI ribbon and  extensible dialog system. Day 2 (Live Meeting 3 hours) 1:30 - 4:30 •         Visual Studio 2010 Tools for SharePoint 2010: Overview of the project and item templates and a walkthrough of the designers •         Sandboxed Solutions: The new deployment model can help mitigate the risk of deploying custom code   •         LINQ to SharePoint:  SharePoint now fully supports LINQ for querying lists Day 3 (Live Meeting 3 hours) 1:30 - 4:30 •         Client Object Model: The Client OM can be accessed via web services, via a client (JavaScript) API, and via REST •         Accessing External Data: Business Connectivity Services (BCS) enables integration with back end systems •         Workflow: A powerful mechanism to create functionality using Windows Workflow Foundation Register for FREE (and tell your colleagues – we have a pretty decent capacity) To take advantage of this you need to: Sign your company up to Microsoft Platform Ready and record your SharePoint interest against one of your companies products Read about Microsoft Platform Ready Navigate to the “Get Technical Benefits” tab for SharePoint and click on Register Today You will then ultimately get an email with details of the Live Meeting to join on the 9th. But you should also favourite the team blog for any last minute details * Such companies are often referred to as an Independent Software Vendors. My team is focused on companies that create products used by many other companies or individuals. That could be a packaged product you can buy "off the shelf" or a Web Site offering a service - the definition is actually pretty wide these days :-) What it does not include is a company building software which will only be used by its own people.

    Read the article

  • When can I publish a software tool written at work?

    - by AlexMA
    I'm working on a software problem at work that is fairly generic, but I can't find a library I like to solve it, so I'm considering writing one myself (at least a bare-bones version). I'll be writing some if not all of the 1.0 version at work, since I need it for the project. If turns out well I might want to bring the work home and polish it up just for fun, and maybe release it as an open-source project. However, I'm concerned that if I wrote the 1.0 version at work I may not be allowed to do this from a legal sense. Obviously I could ask my boss (who probably won't care), but I'm curious how other programmers have dealt with this issue and where the law stands here. My one sentence question is, When is it okay (legally/ethically) to open-source a software tool originally written by you for work at work? What if you have expanded the original source significantly during off-hours? Follow-up: Suppose I write the whole thing at home on my time then simply use it at work, does that change things drastically? Follow-up 2: Note that I'm not trying to rip off my employer (I understand that they're paying me to build products that they own)--I'm just wondering if there's a fair way of doing this for all involved... It would be nice if some nonprofit down the road could use my code and save them some time. Also, there's another issue at stake. If I write the library for a very simple, generic thing (like HTML tables in Javascript), does that mean I can never again do so on my own time without putting myself at legal risk (even if it was a whole new fresh rewrite or a segment of a larger project). Am I surrendering my right to write code for this sort of project for the rest of my life (without this company's permission), since the code at work might still be somewhere in my brain influencing me? This seems related to software patents, as a side-note.

    Read the article

  • In which fields does quality of the software product matter as much as the completion time?

    - by Nav
    Someone told me that if the software product meets clients expectations, it is good quality. But I've worked with Interaction Designers (the same kind of people who made Gmail's interface and usability so cool!), and I've loved working with them because even though they came up with hundreds of changes in requirements, and emphasised on many many subtle details, when the software was complete, I could look at the product and say WOW! The current place I work, the only thing that matters is completing the project on time. As long as it works and as long as the client says it's ok, nobody bothers to improve it. I'm not talking about gold-plating, but I believe that for a programmer to enjoy his (well, maybe her too ;) ) job, they should be able to proudly say that "Hey, I made that software" and that comes only when the product is of good quality. Apart from your opinions on this, I'd also like to know which fields (Eg. Aerospace, Finance etc.) could I find companies (or you could mention the company name) where the quality of a product is as important as completing the project on time?

    Read the article

  • How can we best petition to bring Adobe creative software to Ubuntu?

    - by Sixthlaw
    Now I know its not as simple as asking for Adobe to support their design software on Ubuntu, but is there a way for the community and Canonical to make known to Adobe the rapidly growing amount of Linux users, and their desire for this great set of tools OFFICIALLY? I know that many of the answers I receive might be of the fashion "Its not going to happen", "use the free tools provided" or "Who knows it might happen in the near future", but this IS not what I am looking for. I have noticed, on sites like www.OmgUbuntu.com, there are links to pages where people can "like" the idea. Is there a way to try and get the whole community on board with this one, even Canonical, and as stated above, put forward this proposal to Adobe. The current requests for an Adobe CS, for Linux, are in dribs and drabs scattered all of the internet. Now is the best time to come up with productive solutions on how we can best gather statistics on the amount of people willing to buy the Adobe CS. These are the words of an Adobe employee: "I have forwarded this feedback on to the appropriate team who will consider it for future releases of Adobe software." The larger amount of people we have unified in the ONE community proposal, the greater chance we have of getting the software. How can we make this happen?

    Read the article

  • How can I solve display glitches and poor performance with ATI fglrx driver on my ThinkPad X100e?

    - by rewbs
    Hi all, I noticed that video performance on my Thinkpad X100e was very poor compared to Windows 7, so I installed the ATI fglrx proprietary drivers by using the "Additional Drivers" dialogue box. The system has an ATI Radeon Mobility HD 3200 chip. The result of installing the drivers is pretty devastatingly negative, with symptoms such as skewed content in windows, browser tabs and text boxes failing to refresh when their content changes. In fact, please excuse typos in this post, because I can't really see what I am typing. :) I also notice that HD video playback performance is no better - perhaps even worse - than prior to installing the drivers. Example of what I see: Here's the output of fglrxinfo: display: :0.0 screen: 0 OpenGL vendor string: ATI Technologies Inc. OpenGL renderer string: ATI Radeon HD 3200 Graphics OpenGL version string: 3.3.10237 Compatibility Profile Context Output of lspci | grep -i vga: 01:05.0 VGA compatible controller: ATI Technologies Inc RS780M/RS780MN [Radeon HD 3200 Graphics] I'm on Ubuntu 10.10 with kernel 2.6.35-22-generic-pae. What can I try? Many thanks, -R

    Read the article

  • Is embedded programming closer to electrical engineering or software development?

    - by Jeremy Heiler
    I am being approached with a job for writing embedded C on micro controllers. At first I would have thought that embedding programming is to low on the software stack for me, but maybe I am thinking about it wrong. Normally I would have shrugged off an opportunity to write embedded code, as I don't consider myself an electrical engineer. Is this a bad assumption? Am I able to write interesting and useful software for embedded systems, or will I kick myself for dropping too low on the software stack? I went to school for computer science and really enjoyed writing a compiler, managing concurrent algorithms, designing data structures, and developing frameworks. However, I am currently employed as a Flex developer, which doesn't scream the interesting things I just described. (I currently deal with issues like: "this check box needs to be 4 pixels to the left" and "this date is formatted wrong".) I appreciate everyone's input. I know I have to make the decision for myself, I just would like some clarification on what it means to be a embedded programmer, and if it fits what I find to be interesting.

    Read the article

  • How do I turn off nVidia high performance mode?

    - by gpen06
    I am new to Ubuntu 11.10 and never touched linux until today. I installed Ubuntu alongside with Windows 7. I have an issue with my laptop overheating and freezing when it uses the high performance GPU graphics mode (hybrid card). It is an easy fix for Windows 7. I simply set the graphics mode manually to the low performance integrated graphics (save power) mode using advanced power settings. In Ubuntu, I'm stumped at how to do this. When using Ubuntu, my laptop is showing the same symptoms as it did without the fix for Windows. I have the nVidia graphics driver installed for Windows which allows me to see which mode I am in. I downloaded it off of my laptop's website (ASUS). They do not offer driver downloads for linux. Screenshot of power settings in Windows 7

    Read the article

  • Which metric/list should be used to evaluate whole software development team?

    - by adt
    Title might be seem vague, so let me tell you a little bit history what i am trying to clarify question. I have been hired as a consultant for a corporate's small developement divison ( The company also owns a couple of software dev. companies) My ex manager runs a BI team, with reportes, analyts and developers. He asked me to evaluate overall design, software developement process and code quality . Here what i found, Lots of copy/paste code everywhere ( no reuse ) Even though they have everything TFS, VS Ultimate etc, No Build process , No Cruise Control.net / team city... No unit tests Web Pages with 3700 lines of code, Lots of huge functions ( which can be divided into smaller one's ) No naming convention both db and c# code No 3r party or open source project No IoC No Seperation Of Concerns No Code Quality Check ( NDepend or FxCope or nothing ) No Code Review No Communication within the team They claim they wrote an application framework ( 6 months 3 persons), but I would hardly call a framework ( of course no unit test, there are some but all commented out). Framework contains 14 projects but there are some projects with 1 file 20 lines of code . Honestly, what people are doing fixing bug all thr day( which will provide more bugs eventually), they are kind of isolated from community, some team members even dont know github or stackoverflow they probably went there with google but they dont know about it. So here is question, Is This list ok ? Or am i being picky? Since I dont have any grudge against them, I just want to be fair, honest and I would like to hear you suggestions, before I would submit this list. And since this list also will be review by software division's manager, I dont want any heart break or something like this. http://www.hanselman.com/altnetgeekcode/ For example I would love to such lists, i cant make references. Thanks in advance.

    Read the article

  • How can one find software development work that involves directly the final end user?

    - by RJa
    I've worked in software development for 15 years and, while there have been signficant personal achievements and a lot of experience, I've always felt detached from the man/woman-on-the-street, the every day person, how it affects their lives, in a number of ways: the technologies: embedded software, hidden away, stuff not seen by the everyday person. Or process technology supporting manufactured products the size of the systems, meaning many jobs, divided up, work is abstract, not one person can see the whole picture the organisations: large, with departments dealing with different areas, the software, the hardware, the marketing, the sales, the customer support the locations and hours: out-of-town business parks away from the rest of society, fixed locations, inflexible: 9-5 everyday This to me seems typical of the companies I worked for and see elsewhere. Granted, there are positives such as the technology itself and usually being among high calibre co-workers, but the above points frustrate me about the industry because they detach the work from its meaning. How can one: change these things in an existing job, or compensate for them? find other work that avoids these and connects with the final end user? Job designs tend to focus on the job content and technical requirements rather than how the job aims to fulfil end user needs, is meaningful.

    Read the article

  • Should developers do their own software releases (if there is a prod support team in place)?

    - by leora
    I know there are going to always be differences depending on the particular size, staff etc, but i wanted to get feedback in general around: In an environment where you have a production support team doing first line support and release management, is it better to simply have developers manage their own releases instead? In this case, its internal software at an insurance company but the question should be valid at any company, size, etc I think. Currently, we have our production team do releases but there is an argument that its inefficient and that if you allowed developers the ability to do it, they will focus more on making it simple and efficient and avoid basically passing on scripts, etc to run to another team. The counter argument is that if you don't have a check and balance, you could get a software team (or an individual) that doesn't a very hacky job about getting their software out there (making on the fly changes, not documenting the process, etc) and that by forcing the prod support team to do the actual release, it enforces consistency and proper checks and balances. I know this is not a black or white issue but I wanted to see what folks thought on this so the discipline and consistency is there but without the feeling that an inefficient process is in place.

    Read the article

  • How should a non-IT manager secure the long-term maintenance and development of essential legacy software?

    - by user105977
    I've been hunting for a place to ask this question for quite a while; maybe this is the place, although I'm afraid it's not the kind of "question with an answer" this site would prefer. We are a small, very specialized, benefits administration firm with an extremely useful, robust collection of software, some written in COBOL but most in BASIC. Two full-time consultants have ably maintained and improved this system over more than 30 years. Needless to say they will soon retire. (One of them has been desperate to retire for several years but is loyal to a fault and so hangs on despite her husband's insistence that golf should take priority.) We started down the path of converting to a system developed by one of only three firms in the country that offer the type of software we use. We now feel that although this this firm is theoretically capable of completing the conversion process, they don't have the resources to do so timely, and we have come to believe that they will be unable to offer the kind of service we need to run our business. (There's nothing like being able to set one's own priorities and having the authority to allocate one's resources as one sees fit.) Hardware is not a problem--we are able to emulate very effectively on modern servers. If COBOL and BASIC were modern languages, we'd be willing to take the risk that we could find replacements for our current consultants going forward. It seems like there ought to be a business model for an IT support firm that concentrates on legacy platforms like this and provides the programming and software development talent to support a system like ours, removing from our backs the risks of finding the right programming talent and the job of convincing younger programmers that they can have a productive, rewarding career, in part in an old, non-sexy language like BASIC. Where do I find such firms?

    Read the article

  • How can I write good "research code"?

    - by John
    "Research code" is often held up as a paragon of what not to do when writing software. Certainly, the kind of code that often results from trying to solve a research problem can end up being poorly-designed, difficult to debug, etc. But my question is this: does research code have to be this way? Is it possible to write good research code? Is the only approach to consider the first version a poorly-written prototype to be discarded in favour of the better-designed second version? Software engineering has all sorts of best practices about how to design and write good code, but I don't usually find this relevant when you don't have a good idea ahead of time what the final system will look like. The final system is likely to be a result of what did or didn't work along the way, and the only way to determine what does or doesn't work is to write the code first. As you find things that don't work, you change what the final system looks like, moving further away from your original design (assuming you had one). I'd be interested in any personal experience with these issues, as well as any books or other resources anyone can point me to.

    Read the article

< Previous Page | 280 281 282 283 284 285 286 287 288 289 290 291  | Next Page >