Was I asking you? OK... there's only a handful of people in ET who realizes the issue with QuickFIX's design. Especially, from the code I posted. ... I'm going to be very protective/defensive... 1. It's a copy/paste of the prototype of what I've written 3 years ago. Considering that I know nitro in person, I tried hiring the dude for a low 6 digit but he refused. Now, I can see that I can't afford to hire him for his true intrinsic value. (Even though, he sucks at selling himself) The reason is he's a top notch Financial IT. No more to add, except that I appreciate his posts/contribution in ET. Seriously, last thing I want is an expert Financial IT replying to my posts. That said, his reply is complete relevant because he's extremely experienced regarding this matter. Seriously, there's only a handful of people who has the potential to conclude with QuickFIX's design flaws from my codes. 2. The post I made was done based on the fact that the OP is a junior IT. I expected him to be naive about general FIX specifications and QuickFIX. Even if the OP came head-on, I was planning on going into FAST and FIX 5.0 but it's now irrelevant (C'mon... I've been in ET for quite sometime... I can expect the competency and their answers I get from people based on their original posts... hence I'm look at like a troll and flamer...) 3. My final conclusion, which I was trying to get at with my planned series of post, was that the dude(OP) is missing the point regarding his reply towards rosy2. It's not like school where your professor tells you what to do. In a professional environment, you're given a "subjective" task and you have a bunch of people criticizing you about how you approached it (and that's how one learns to work in a professional environment). Anyways... nitro... thanks for spoiling my planned series of contribution. I'll end my post in this thread here... I hate posting in your thread, you get me good... Considering the OP, it may not have helped him by overwhelming the reply. But it help ET's ATS forums... I really don't know... really... I don't know if your feedback is good...
Anyone competent can see, QuickFix experience or not, that the posted code is pretty horrible...and not just because of QuickFix. There's a lot of duplicate and boilerplate code. Boilerplate code normally sets alarm bells ringing for me. I'm sure with a little bit of thought, factory methods etc. things could be engineered a whole lot better. Why there is so much commented code is beyond me. Ironic given that the code is so anemic in terms of real "comments". Code like: if (status == true) makes me cringe. In short, I don't think a decent coder wouldn't have written the code that way in the first place...but I'm no expert.
Why the heck does it matter that you can reduce some legacy code to 1/5 (in number of lines of code)?! If it is well encapsulated, has got a usable/sensible interface - then it is if it aint broken, dont try to fix it. Performance issues most likely lie elsewhere than in a low level library like this. (Although - libraries that write debugging / log messages to the console are annoying.)
Ok first of all, what makes me an idiot out of everything I said? Second of all, did I not state very clearly that I am a junior developer that knows c++, java, and c#, did you see me write I am an experienced trading developer anywhere? I have never worked with QuickFix in my life but I will gladly do what you are asking me to do but you need to give me more details.... Sorry to say, but its people like you that give programmers a bad name, just rude and arrogant for no reason at all.... You think by calling me an idiot and posting a chunk of code that is meaningless to me, you have displayed how good a programmer you are.... Thats ridiculous and furthermore your code that you posted is unacceptable even by college standards... Almost no comments or line docs in the code, and its full of commented out lines of code.... its structure just looks very disorganized.... I maybe junior but I have worked in the field and by your work that you are proud to advertise, I really doubt you are a top notch developer, just my opinion. And furthermore what you are doing is just like me coming to you and giving you a chunk of code used by a mobile device (if you have never programmed one before) and not tell you any thing about its specific purpose or any specs at all, then tell you to optimize it for me.... If you were a real developer you should know there is more to development than just knowing how to code in a specific language... Just cause I know C# doesn't mean you can just throw anything at me with no details and tell me to make it better for you. What I meant by give me a chunk of code and ask me other ways to write it are basic things like using different types of collections to do the same thing or different ways to create threads etc... But no worries, if you are serious about testing me, then give me more specs about the code and what you meant for it to do then tell me to optimize it for you and I will gladly do it and I'll even throw in a bonus and make your code look organized for you.
Here is something useful to know, and a typical project that you are handed. http://www.iseoptions.com/WebForm/viewPage.aspx?categoryId=246&header2=true&menu3=true&link0=true Scroll down to "Working with the OPRA FAST Decoder" Maybe read the pdf file above under "OPRA Message Formats and Encoding" Download the Linux version first. Get it to compile. Make sure you follow the instructions. Run it on the demo data they provide. It core dumps. Get it to work. Show what you did to get it to work. Then, do the same thing for the windows code. Post back here. You may get a job out of it from someone watching, but even if you don't you will have gained some interesting experience. This sort of work is worth $150k in Chicago and double that in NYC.
You are a very funny person honestly. Do you know the difference between a junior level person and an entry level person? I never said I was straight out of college so please keep the differentiating between college and real world to yourself, right out of school I was programming for a big 4 consulting firm so I doubt I am oblivious to the difference between school and real world. Furthermore what I was trying to emphasize to rosy2 is that what she was asking of me was not an entirely a good way of testing somebody. First of all I could easily get another person's code and say its mine, and lets even leave that case out for now, but realistically what code could I really show her? Do you want me to get code from work and post it here so I can get fired or better yet sued? Or do you want me to get code from personal projects I done for people that have paid me for ownership of the code? So since I did not keep any code from school, if I say I don't have code to post, does it mean I am lying about being a good programmer? So what I told rosy2 about that not being the best way to test me was on point.... So instead of being cynical, just respond to the topic I posted originally, cause I didn't come here to flame people or get flamed, I am just looking for constructive opinions cause I am new to this field but I am trying to get in and I am humbly asking for all the guidance help I can get....
I feel for you, man. Let me say, though, that the interviews you want are just as brutal as the ET threads. The people may be different, but the attitudes are the same. It's just the world we live in. Since we are both in the job market together, perhaps we should exchange IMs or something. I just took one of those Meyers-Briggs personality tests for a job application. I usually get ESFP. However, in the software world, most people are INTJ or INTP. These are not empathetic people and a lot of them have Aspberger's syndrome. You have to think of these things in the back of your mind and just not take it very personally.
1. I had a six pack inside me when I made that post. I'm not going to deny that I was drunk. 2. I asked you to find flaws in the code. Do you really expect me to post a "perfect code"? It was posted because it was flawed. Yeah... I was being a bit to cynical. I admit that I was an A-hole, but that doesn't mean I'm going to stop drinking.
I wont spoil for kci but I'm starting to think I'm in the wrong field for being able to track down simple C idiosyncrasies..