Search Results

Search found 4946 results on 198 pages for 'proper benchmarking'.

Page 51/198 | < Previous Page | 47 48 49 50 51 52 53 54 55 56 57 58  | Next Page >

• Beginner's guide to writing comments?

  - by Cameron

  Is there a definitive guide to writing code comments, aimed at budding developers? Ideally, it would cover when comments should (and should not) be used, and what comments should contain. This answer: Do not comment WHAT you are doing, but WHY you are doing it. The WHAT is taken care of by clean, readable and simple code with proper choice of variable names to support it. Comments show a higher level structure to the code that can't be (or is hard to) show by the code itself. comes close, but it's a little concise for inexperienced programmers (an expansion on that with several examples and corner cases would be excellent, I think).

  Read the article

• How to write reusable code in node.js

  - by lortabac

  I am trying to understand how to design node.js applications, but it seems there is something I can't grasp about asynchronous programming. Let's say my application needs to access a database. In a synchronous environment I would implement a data access class with a read() method, returning an associative array. In node.js, because code is executed asynchronously, this method can't return a value, so, after execution, it will have to "do" something as a side effect. It will then contain some code which does something else than just reading data. Let's suppose I want to call this method multiple times, each time with a different success callback. Since the callback is included in the method itself, I can't find a clean way to do this without either duplicating the method or specifying all possible callbacks in a long switch statement. What is the proper way to handle this problem? Am I approaching it the wrong way?

  Read the article

• How should one refer to their social network profiles in prints?

  - by Omne

  In case a person has many different social network profiles, what is the proper way of referring to those profiles in a work that is meant to be printed or may be viewed as an image? For example one owns the username "CoolCompany" on Facebook, Twitter, Google+ and other websites. Should they provide a URL for each profile? for example: Facebook.com/CoolCompany Twitter.com/CoolCompany ... I often notice that in videos or on TV people may only use the logos but I believe it is not very useful, specially in case that the audience are unfamiliar with a social network... Is it acceptable to provide the logos or names of the networks and mention the username only once like this: Facebook & Twitter: @CoolCompany Or Facebook & Twitter: /CoolCompany

  Read the article

• Oneiric desktop looks seriously defect after updates

  - by Matthijs

  Probably a coincidence, but after applying the fix to How to get the proper LightDM theme? and running the latest batch of updates, my desktop suddenly looked like this after rebooting: Funny thing is, it seems to be account specific, as my girlfriend's account seems unaffected. I can't logout because the whole entry is missing on the top right. Rebooting doesn't solve the issue. System settings (when run from the launcher) lacks a whole lot of icons and changeing the theme doesn't solve the problem (only changes the window top bar). I'm going to try the solution in How do I fix my theme? first, but if that doesn't work (likely, since I seem to have a slightly bigger problem than the one in that question) I'd love something else to try (other than creating a new account).

  Read the article

• What is a best practice tier structure of a Java EE 6/7 application?

  - by James Drinkard

  I was attempting to find a best practice for modeling the tiers in a Java EE application yesterday and couldn't come up with anything current. In the past, say java 1.4, it was four tiers: Presentation Tier Web Tier Business Logic Tier DAL (Data Access Layer ) which I always considered a tier and not a layer. After working with Web Services and SOA I thought to add in a services tier, but that may fall under 3. the business logic tier. I did searches for quite a while and reading articles. It seems like Domain Driven Design is becoming more popular, but I couldn't find a diagram on it's tier structure. Anyone have ideas or diagrams on what the proper tier structure is for newer Java EE applications or is it really the same, but more items are ranked under the four I've mentioned?

  Read the article

• Topics for development team cross training sessions

  - by BBlake

  Our team of developers are going to start holding monthly meetings for the purposes of cross training and knowledge improvement. We're looking for ideas for topics to discuss. We've already made a list of some obvious ones, such as discussions/training on specific applications, proper usage of TFS for source control, bug tracking and code reviews, coding standards, and corporate architecture. The problem we're having is that we are a cross-platform development team so we don't want to look at topics that only apply to certain members of the team (Sql, .NET, reporting, third party apps, etc). We'll use sub-team meetings for those. So what other topics that would apply across a broad development team would be good for these training sessions?

  Read the article

• Rebind Alt key to win using setxkbmap?

  - by Wayne Werner

  Hi, After an hour or two of manpage and Google searching and finding no solution or good resources, I've come for help! I have set my Caps Lock key to Ctrl using setxkbmap -option ctrl:nocaps - this works perfectly fine. However, since I use [awesome][1], and an IBM model M which lacks the meta key, I need my left alt key to replace the windows key. Using xkeycaps I was able to get this to work, except it killed my arrow keys and End. Problematic. Unfortunately, documentation on setxkbmap options are sparse. and I can't find the proper option to use. Thanks for any links/solutions.

  Read the article

• Client Side Form Validation vs. Server Side Form Validation

  In my opinion, it is mandatory to validate data using client side and server side validation as a fail over process. The client side validation allows users to correct any error before they are sent to the web server for processing, and this allows for an immediate response back to the user regarding data that is not correct or in the proper format that is desired. In addition, this prevents unnecessary interaction between the user and the web server and will free up the server over time compared to doing only server side validation. Server validation is the last line of defense when it comes to validation because you can check to ensure the user’s data is correct before it is used in a business process or stored to a database. Honestly, I cannot foresee a scenario where I would only want to use one form of validation over another especially with the current cost of creating and maintaining data. In my opinion, the redundant validation is well worth the overhead.

  Read the article

• No app icons in menus

  - by Lyrositor

  I have enabled icons in menus, and most of them show up well (folder icons, file icons, save icons, etc.). But whenever an app tries to use an icon specific to itself, such as GIMP, it replaces that icon with the app icon, so that when I look at a menu in GIMP, I see most of the time just icons of GIMP's logo (that seems to be the only problem; the text is fine). I thought this was proper to GIMP, but I also noticed it happening on AbiWord. Thankfully, as I said, it's not for all icons. But it still bugs me. Why is it so? Here is a screenshot of a menu in GIMP: I am using Ubuntu 11.04 Natty Narwhal, with Unity desktop (the problem also occurs on the Classic desktop, if I remember correctly). EDIT: This does NOT occur on the Classic desktop; only in Unity.

  Read the article

• Best exception handling practices or recommendations?

  - by user828584

  I think the two main problems with my programs are my code structure/organization and my error handling. I'm reading Code Complete 2, but I need something to read for working with potential problems. For example, on a website, if something can only happen if the user tampers with data via javascript, do you write for that? Also, when do you not catch errors? When you write a class that expects a string and an int as input, and they aren't a string and int, do you check for that, or do you let it bubble up to the calling method that passed incorrect parameters? I know this is a broad topic that can't be answered in a single answer here, so what I'm looking for is a book or resource that's commonly accepted as teaching proper exception handling practice.

  Read the article

• Is it OK to have multiple asserts in a single unit test?

  - by Restuta

  I think that there are some cases when multiple assertions are needed (e.g. Guard Assertion), but in general I try to avoid this. What is your opinion? Please provide a real word examples when multiple asserts are really needed. Thanks! Edit In the comment to this great post Roy Osherove pointed to the OAPT project that is designed to run each assert in a single test. This is written on projects home page: Proper unit tests should fail for exactly one reason, that’s why you should be using one assert per unit test. And also Roy wrote in comments: My guideline is usually that you test one logical CONCEPT per test. you can have multiple asserts on the same object. they will usually be the same concept being tested.

  Read the article

• Best practices for managing deployment of code from dev to production servers?

  - by crosenblum

  I am hoping to find an easy tool or method, that allow's managing our code deployment. Here are the features I hope this solution has: Either web-based or batch file, that given a list of files, will communicate to our production server, to backup those files in different folders, and zip them and put them in a backup code folder. Then it records the name, date/time, and purpose of the deployment. Then it sends the files to their proper spot on the production server. I don't want too complex an interface to doing the deployment's because then they might never use it. Or is what I am asking for too unrealistic? I just know that my self-discipline isn't perfect, and I'd rather have a tool I can rely on to do what needs to be done, then my own memory of what exact steps I have to take every time. How do you guys, make sure everything get's deployed correctly, and have easy rollback in case of any mistakes?

  Read the article

• How to avoid Hotmail/Live rejections for (legit) large volume eMailing?

  - by vmarquez

  While qualifying eMail for Spam, Hotmail/Live checks the historical records of numbers of eMails sent by a sender (FROM, eMail Server, IP, etc.). Some times, perfectly valid bulk eMails that are not Spam, (i.e. double opt-in list, from a server with proper SPF Record, signed with DKIM, unregister links and contact info, etc.) are rejected and not delivered to destinataries. Not even to their Junk folder. I guess we can avoid this situation by progressivelly "training" Hotmail/Live about the reputation of our sender and sending small quantyties of eMails innitially and increasing the quantity for some amount/percentaje during each delivery. Are there guidelines or do you have any experience on these quantities, strategy, solutions? Thank you in advance. EDIT: This question with a bounty is still unanswered. 8 hours to be automatically awarded! Do you have the answer?

  Read the article

• Berkeley DB Java Edition 4.0.103 Available

  - by charles.lamb

  We'd like to let you know that JE 4.0.103 is now at http://www.oracle.com/technology/software/products/berkeley-db/je/index.html. The patch release contains both small features and bug fixes, many of which were prompted by feedback on this forum. Some items to note: New CacheMode values for more control over cache policies, and new statistics to enable better interpretation of caching behavior. These are just one initial part of our continuing work in progress to make JE caching more efficient. Fixes for proper cache utilization calculations when using the -XX:+UseCompressedOops JVM option. A variety of other bug fixes. There is no file format or API changes. As always, we encourage users to move promptly to this new release.

  Read the article

• Broadcom STA driver doesn't work well with BCM4313

  - by Oli

  Following on from my other question about our new Samsung Q330, I've noticed that the wireless is incredibly flakey. It can connect but after a little use, especially if it does a lot of downloading at once (read: install something from the Software Centre), the connection stops working. Network Manager still see the connection, there's just no network throughput. I've simple tests like pinging other local network hosts and they just fail. The Samsung Q330 has a Broadcom BCM4313 wifi card (it's proper ID: 14e4:4727) and it's running on the Broadcom STA drivers that Jockey suggests (it didn't work at all without this). I did try installing b43-fwcutter but this just didn't do anything. I was expecting a configuration screen to come up (to select a firmware) but it never did. This page suggests the newer brcm80211 driver might be able to help, but I don't know how to install that. If you think this is the right route, please let me know how one goes about installing it.

  Read the article

• Why is my wi-fi STILL not working?

  - by Alex Grantham

  i have 10.10, and have gotten ndiswrapper and the proper driver, it installed properly, but still my wi-fi isn't working. In terminal it say its disabled, but its enabled EVERYWHERE... even when I exited windows. Right now i'm left to use an Ethernet cord and a second computer... please help. My windows isn't in good condition and I would love to switch to Ubuntu... but wi-fi is definitely a need. FYI, I have a toshiba satellite a505d-s6958 with the realtek rtl819[somehting something something]... but I have the driver from their webisite... I really just need to get my wifi to work

  Read the article

• What will be the impact on SEO if we remove our SSL certificate (url become http instead of https)?

  - by pixeline

  For some weird reason, our domain's content is returned for any https request set to any of our server's hosted domain names. https://domain.com leads to our website, with a proper SSL certificate (so, no warning). https://domain2.com, also hosted on our server but without SSL certificate, leads to a warning, and if accepted, to our website's content! The problem is that any search for our keywords in Google shows "fake websites" on top of ours, with the warning et al. It seems unsolvable so we are thinking about switching back ton nonsecure http . I'm just afraid of losing whatever indexing we have. How can i avoid that? Thanks, a.

  Read the article

• Should I demand unit-testing from programmers?

  - by Morten

  I work at a place, where we buy a lot of IT-projects. We are currently producing a standard for systems-requirements for the requisition of future projects. In that process, We are discussing whether or not we can demand automated unit testing from our suppliers. I firmly believe, that proper automated unit-testing is the only way to document the quality and stability of the code. Everyone else seems to think that unit-testing is an optional method that concerns the supplier alone. Thus, we will make no demands of automated unit-testing, continous testing, coverage-reports, inspections of unit-tests or any of the kind. I find this policy extremely frustrating. Am I totally out of line here? Please provide me with arguments for any of the oppinions.

  Read the article

• Simulating simultaneous entities

  - by Steven Jeuris

  Consider the need to simulate a set of entitities in an accurate way. All entities exist in an artificial timeline. Within 'steps' of this timeline, all entities can do certain operations. It is imperative that timed events, are handled accurately, and not in processing order. So simple threading isn't a proper simulation, nor is procedurally walking across all entities. Processing may be slow, accuracy is key here. I have some ideas how to implement this myself, but most likely something like this has been done before. Are there any frameworks available for these purposes? Is there any particular paradigm more suitable?

  Read the article

• Game-a-Week One Completed

  - by Matt Christian

  Last night I finished my Game-a-Week One and felt and extreme sense of accomplishment with what I finished in a single week. I removed all traces of the JigLibX code since it wasn't working properly due to my implementation and got my collision working thanks to some BoundingSpheres and Riemer's tutorials.  However, since the characters are Corndogs a rectangle bounding box would have made more sense although the bullets are only able to move forward currently. While developing it in a week was a challenge, developing it in a week while maintaining proper coding standards and clean, reusable code was damn near impossible.  It's possible my next step will be to either refactor it or move onto Game-a-Week Two. I will post a link to the game ZIP in the future.

  Read the article

• Is it possible to construct a cube with fewer than 24 vertices

  - by Telanor

  I have a cube-based world like Minecraft and I'm wondering if there's a way to construct a cube with fewer than 24 vertices so I can reduce memory usage. It doesn't seem possible to me for 2 reasons: the normals wouldn't come out right and per-face textures wouldn't work. Is this the case or am I wrong? Maybe there's some fancy new DX11 tech that can help? Edit: Just to clarify, I have 2 requirements: I need surface normals for each cube face in order to do proper lighting and I need a way to address a different indexes in a texture array for each cube face

  Read the article

• Does having multiple URIs mapping to the same resource help SEO?

  - by Brian Wheeler

  Let's say I have a site with products that have tags, if each resource is available at GET '/products/tagged/:tag_list/:product_permalink' Could that be better for SEO than just one permalink? For example a product tagged "tea" and "coffee" would be available at GET '/products/tagged/tea/:product_permalink' GET '/products/tagged/coffee/:product_permalink' GET '/products/tagged/tea/coffee/:product_permalink' GET '/products/tagged/coffee/tea/:product_permalink' I would imagine that google would appreciate this because it gives multiple URIs with different levels of detail about the product, but I cant really be certain. Anyone have any direct knowledge on the topic? --EDIT-- As John Conde points, this is a horrible idea. What about having the links on my site link to a route such as GET '/products/tagged/:full_tag_list/:product_permalink', and then any time a user changes tags just have a HTTP moved permanently status to the new URL. Therefore duplicate URLs would be highly unlikely and mitigated by the proper response. Would this be better?

  Read the article

• How to Install Nvidia Drivers

  - by Richard Rodriguez

  I just ordered the Nvidia GTX 560 card, which should arrive tomorrow. I have a dilemma, though. Should I keep using the driver which is available in "additional drivers" in Ubuntu (10.10), or should I install the driver from the nvidia site? NOTE - The methods to install explained here apply to all Nvidia, Ati & Intel video cards The latest driver available at the nvidia site: LINUX X64 (AMD64/EM64T) DISPLAY DRIVER Version: 280.13 Certified Release Date: 2011.08.01 Operating System: Linux 64-bit Language: English (U.S.) File Size: 52.4 MB I should point out that I don't need the card to unleash its full potential in Ubuntu (I have Windows for gaming, other HDD), I just need it to work properly, that meaning the power saving should work (I don't want the card to overheat for no reason), also I would like the fans to work at proper speeds, etc. So which driver is the best for me?

  Read the article

• How to setup AD group to work with SharePoint group security

  - by ybbest

  Issue: Recently, I am having issues with the permission setup at client side. The way we setup permission is we create AD Group and add the add AD Group to the proper SharePoint Group. If we need to grant permission to a user, we then add the user to the correct AD group. But end user reports that by adding a user to an AD group, this user does not have any access to the SharePoint site. Solution: Change the AD Group scope from Global to Universal fix the problem. From To References: AD Group Scopes from Paul Stork http://social.technet.microsoft.com/forums/en-US/sharepointadminprevious/thread/79d2af40-3daa-4f61-86e5-5bb54086147f

  Read the article

• How do you make comp.sci students and future programmers aware of the various software licenses and the nuances of it ?

  - by Samyak Bhuta

  To be specific How would you include it as part of curriculum ? Would it be too boring to just introduce them as a pure law subject ? Are there any course structure available or can we derive one ? What are the books that could be used ? I would like to see that - after going through the course - candidate is well aware of "what software licenses are and what they are good for". Various implications of not knowing it in it's proper sense. What licenses they should use for their own code. What to consider when they are trying to use certain libraries or tools in their project and gauge risks/rewards associated with it. The idea is to let them make informed choices when they are professionals/practitioners in field of programming and not make them substitute for a lawyer or even a paralegal who is going to fight the case or draft things.

  Read the article

< Previous Page | 47 48 49 50 51 52 53 54 55 56 57 58  | Next Page >