Search Results

Search found 16427 results on 658 pages for 'brendan long'.

Page 198/658 | < Previous Page | 194 195 196 197 198 199 200 201 202 203 204 205  | Next Page >

  • Code review vs pair programming

    - by mericano1
    I was wondering what is the general idea about code review and pair programming. I do have my own opinion but I'd like to hear from somebody else as well. Here are a few questions, please give me your opinion even on some of the point First of all are you aware of way to measure the effectiveness of this practices? Do you think that if you pair program, code reviews are not necessary or it's still good to have them both? Do you think anybody can do code review or maybe is better done by seniors only? In terms of productivity do you think it suffers from pairing all the times or you will eventually get in back in the long run? Thanks!

    Read the article

  • Career Advice: masters degree or work experience [on hold]

    - by user95488
    I graduated about four years ago with a degree in mathematics and I currently work as a Software Developer/Business Analyst, but more Business Analyst. I've been working for about 4 years and and I am concerned with my long term career path. I would strongly prefer to do much more software development but to continue on my current path would lead me to an analyst role. I was thinking of getting a masters in computer science to help refocus my career toward core software development. Does anyone have any advice here? Is this a bad idea?

    Read the article

  • When designing a job queue, what should determine the scope of a job?

    - by Stuart Pegg
    We've got a job queue system that'll cheerfully process any kind of job given to it. We intend to use it to process jobs that each contain 2 tasks: Job (Pass information from one server to another) Fetch task (get the data, slowly) Send task (send the data, comparatively quickly) The difficulty we're having is that we don't know whether to break the tasks into separate jobs, or process the job in one go. Are there any best practices or useful references on this subject? Is there some obvious benefit to a method that we're missing? So far we can see these benefits for each method: Split Job lease length reflects job length: Rather than total of two Finer granularity on recovery: If we lose outgoing connectivity we can tell them all to retry The starting state of the second task is saved to job history: Helps with debugging (although similar logging could be added in single task method) Single Single job to be scheduled: Less processing overhead Data not stale on recovery: If the outgoing downtime is quite long, the pending Send jobs could be outdated

    Read the article

  • HP 655 Notebook (Ubuntu 12.04) wireless internet disabled after running updates

    - by Bastian
    Today the update-manaeger suggested that I should download some updates. So I did. After the updates were installed I had to reboot the system. After the reboot my notebook doesn't see any wireless networks. I was looking for a long time on the internet for answers but non of them applies to my case. I have the idea, it has got something to do with the kernel I am currently using (Linux 3.2.0-55). Is this one new for Ubuntu 12.04? When I boot my system I can choose to use an older kernel (linux 3.2.0-32). When I use this one the notebook does see the wireless networks. This is my network card according to the command lspci: Ralink corp. RT3290 Wireless 802.11n 1T/1R PCIe Anyone an idea to fix my problems with wireless internet?

    Read the article

  • Speaking at PASS 2012 Summit in Seattle #sqlpass

    - by Marco Russo (SQLBI)
    I will deliver two sessions at the next PASS Summit 2012: one is title Inside DAX Query Plans and the other is Near Real-Time Analytics with xVelocity (without DirectQuery).These will be two sessions that require a lot of preparation and even if I have already much to say, I still have a long work to do this summer in order to go deeper in several details that I want to investigate for completing these sessions.I already look forward to come back in Seattle!In the meantime, you have to study SSAS Tabular and if you want to get a real jumpstart why not attending one of the next SSAS Tabular Workshop Online? We are working on more dates for this fall, but there are a few dates already scheduled.And, last but not least, the early Rough Cuts edition of our upcoming SSAS Tabular book is finally available here (really near to the final print)!

    Read the article

  • Windows 7/Ubuntu 12 dual boot deleted for Windows 8 installation. How to make grub rescue go away?

    - by dimious
    I had a Windows 7/Ubuntu 12 dual boot and I decided to clean install Windows 8 over them. The problem is that after I deleted all partitions and installed windows I was getting an "Operation system not found", however after an "enter" the system will normally boot into Windows 8. I realized that Windows did their trick and put the system (not partition anymore?!?) "tag" (Disk Management) on my media hard drive. After trying to fix the boot/mbr to be able to boot from my main drive the "Operation system not found" changed to the "grub rescue" prompt. I know that I cannot use that because I have killed the grub files. Windows can still boot as long as I choose to boot from the media drive. The question is, is there any way to move the "system", whatever it is now, to the main drive and have the PC boot from there, while making grub disappear? And if that is possible after that, can I just make the Media drive inactive or I will have to somehow remove the "system" tag?

    Read the article

  • How far back do you use your version control and for what reason?

    - by acidzombie24
    Typically when i work on a project i only go back a few days or the last major change when i decide to do something drastic. I sometimes notice i broke a test or a feature and overlooked it for a few weeks so i may go back a month or two and see if the feature or test is broken and trace down the week i broke it. Then find what change did it. On a long term project over the span of a year. Do you actually go back 6+ months and if so why?

    Read the article

  • Ubuntu 11.10 momentarily freezes every couple minutes

    - by Jeff Welling
    I'm working on an Acer Aspire One D257 running Ubuntu 11.10, and every few minutes the laptop freezes up for a second or two, before becoming responsive again. This doesn't really present a huge problem, but when I'm trying to code all of a sudden vim stops responding for a couple seconds while I'm trying to modify my code, it's increasingly frustrating. The odd part is, if I play a movie in VLC at the same time, when vim pauses the video keeps playing just fine (VLC's priority is not modified). I'm wondering if anyone knows why this may be happening, or in lieu of that, how I might be able to track down the source of my frustrating pauses. Normally I would try using top, but the pauses are shorter than 2 seconds so nothing unusual shows up in top as consuming the CPU. Does anyone know how to troubleshoot intermittent repeating 1 second long pauses in vim?

    Read the article

  • Developing Schema Compare for Oracle (Part 4): Script Configuration

    - by Simon Cooper
    If you've had a chance to play around with the Schema Compare for Oracle beta, you may have come across this screen in the synchronization wizard: This screen is one of the few screens that, along with the project configuration form, doesn't come from SQL Compare. This screen was designed to solve a couple of issues that, although aren't specific to Oracle, are much more of a problem than on SQL Server: Datatype conversions and NOT NULL columns. 1. Datatype conversions SQL Server is generally quite forgiving when it comes to datatype conversions using ALTER TABLE. For example, you can convert from a VARCHAR to INT using ALTER TABLE as long as all the character values are parsable as integers. Oracle, on the other hand, only allows ALTER TABLE conversions that don't change the internal data format. Essentially, every change that requires an actual datatype conversion has to be done using a rebuild with a conversion function. That's OK, as we can simply hard-code the various conversion functions for the valid datatype conversions and insert those into the rebuild SELECT list. However, as there always is with Oracle, there's a catch. Have a look at the NUMTODSINTERVAL function. As well as specifying the value (or column) to convert, you have to specify an interval_unit, which tells oracle how to interpret the input number. We can't hardcode a default for this parameter, as it is entirely dependent on the user's data context! So, in order to convert NUMBER to INTERVAL DAY TO SECOND/INTERVAL YEAR TO MONTH, we need to have feedback from the user as to what to put in this parameter while we're generating the sync script - this requires a new step in the engine action/script generation to insert these values into the script, as well as new UI to allow the user to specify these values in a sensible fashion. In implementing the engine and UI infrastructure to allow this it made much more sense to implement it for any rebuild datatype conversion, not just NUMBER to INTERVALs. For conversions which we can do, we pre-fill the 'value' box with the appropriate function from the documentation. The user can also type in arbitary SQL expressions, which allows the user to specify optional format parameters for the relevant conversion functions, or indeed call their own functions to convert between values that don't have a built-in conversion defined. As the value gets inserted as-is into the rebuild SELECT list, any expression that is valid in that context can be specified as the conversion value. 2. NOT NULL columns Another problem that is solved by the new step in the sync wizard is adding a NOT NULL column to a table. If the table contains data (as most database tables do), you can't just add a NOT NULL column, as Oracle doesn't know what value to put in the new column for existing rows - the DDL statement will fail. There are actually 3 separate scenarios for this problem that have separate solutions within the engine: Adding a NOT NULL column to a table without a rebuild Here, the workaround is to add a column default with an appropriate value to the column you're adding: ALTER TABLE tbl1 ADD newcol NUMBER DEFAULT <value> NOT NULL; Note, however, there is something to bear in mind about this solution; once specified on a column, a default cannot be removed. To 'remove' a default from a column you change it to have a default of NULL, hence there's code in the engine to treat a NULL default the same as no default at all. Adding a NOT NULL column to a table, where a separate change forced a table rebuild Fortunately, in this case, a column default is not required - we can simply insert the default value into the rebuild SELECT clause. Changing an existing NULL to a NOT NULL column To implement this, we run an UPDATE command before the ALTER TABLE to change all the NULLs in the column to the required default value. For all three, we need some way of allowing the user to specify a default value to use instead of NULL; as this is essentially the same problem as datatype conversion (inserting values into the sync script), we can re-use the UI and engine implementation of datatype conversion values. We also provide the option to alter the new column to allow NULLs, or to ignore the problem completely. Note that there is the same (long-running) problem in SQL Compare, but it is much more of an issue in Oracle as you cannot easily roll back executed DDL statements if the script fails at some point during execution. Furthermore, the engine of SQL Compare is far less conducive to inserting user-supplied values into the generated script. As we're writing the Schema Compare engine from scratch, we used what we learnt from the SQL Compare engine and designed it to be far more modular, which makes inserting procedures like this much easier.

    Read the article

  • Sponsored Giveaway: Free Copies of WinX DVD Ripper Platinum for All How-To Geek Readers

    - by The Geek
    Have you ever wanted to watch a movie on your iPad, iPhone, Android tablet, or even your computer… without having to pay to download it from iTunes? You can easily convert DVDs to digital formats using WinX DVD Ripper Platinum, and we’re giving away free copies to all How-To Geek readers. To get your free copy, just click through the following link to download and get the license code, as long as you download it by November 27th. For Windows users: http://winxdvd.com/giveaway/ For Mac users: http://www.macxdvd.com/giveaway/giveaway.htm   Why Does 64-Bit Windows Need a Separate “Program Files (x86)” Folder? Why Your Android Phone Isn’t Getting Operating System Updates and What You Can Do About It How To Delete, Move, or Rename Locked Files in Windows

    Read the article

  • Immutable design with an ORM: How are sessions managed?

    - by Programmin Tool
    If I were to make a site with a mutable language like C# and use NHibernate, I would normally approach sessions with the idea of making them as create only when needed and dispose at request end. This has helped with keeping a session for multiple transactions by a user but keep it from staying open too long where the state might be corrupted. In an immutable system, like F#, I would think I shouldn't do this because it supposes that a single session could be updated constantly by any number of inserts/updates/deletes/ect... I'm not against the "using" solution since I would think that connecting pooling will help cut down on the cost of connecting every time, but I don't know if all database systems do connection pooling. It just seems like there should be a better way that doesn't compromise the immutability goal. Should I just do a simple "using" block per transaction or is there a better pattern for this?

    Read the article

  • Is deserializing complex objects instead of creating them a good idea, in test setup?

    - by Chris Bye
    I'm writing tests for a component that takes very complex objects as input. These tests are mixes of tests against already existing components, and test-first tests for new features. Instead of re-creating my input objects (this would be a large chunk of code) or reading one from our data store, I had the thought to serialize a live instance of one of these objects, and just deserialize it into test setup. I can't decide if this is a reasonable idea that will save effort in long run, or whether it's the worst idea that I've ever had, causing those that will maintain this code will hunt me down as soon as they read it. Is deserialization of inputs a valid means of test setup in some cases? To give a sense of scale of what I'm dealing with, the size of serialization output for one of these input objects is 93KB. Obtained by, in C#: new BinaryFormatter().Serialize((Stream)fileStream, myObject);

    Read the article

  • Is there a media player that works on HTTPS sites?

    - by Iain Hallam
    I'm currently using Yahoo! Media Player for a site that needs to play MP3 files that are stored on our server. In total, there's quite a bit more than the free limits at Soundcloud, but each file is only a few minutes long. YMP is pretty good, but causes security warnings on HTTPS pages, because it can only be served via HTTP. Is there an equivalent free player I can embed for the HTTPS pages? EDIT: Just to clarify, I'm initially looking for something that will scan the page and turn media links playable.

    Read the article

  • World Backup Day

    - by red(at)work
    Here at Red Gate Towers, the SQL Backup development team have been hunkered down in their shed for the last few months, with the toolbox, blowtorch and chamois leather out, upgrading SQL Backup. When we started, autumn leaves were falling. Now we're about to finish, spring flowers are budding. If not quite a gleaming new machine, at the very least a familiar, reliable engine with some shiny new bits on it will trundle magnificently out of the workshop. One of the interesting things I've noticed about working on software development teams is that the team is together for so long 'implementing' stuff - designing, coding, testing, fixing bugs and so on - that you occasionally forget why you're doing what you're doing. Doubt creeps in. It feels like a long time since we launched this project in a fanfare of optimism and enthusiasm, and all that clarity of purpose and mission "yee-haw" has dissipated with the daily pressures of development. Every now and again, we look up from our bunker and notice all those thousands of users out there, with their different configurations and working practices and each with their own set of problems and requirements, and we ask ourselves "does anyone care about what we're doing?" Has the world moved on while we've been busy? Could we have been doing something more useful with the time and talent of all these excellent people we've assembled? In truth, you can research and test and validate all you like, but you never really know if you've done the right thing (or at least, something valuable for some users) until you release. All projects suffer this insecurity. If they don't, maybe you're not worrying enough about what you're building. The two enemies of software development are certainty and complacency. Oh, and of course, rival teams with Nerf guns. The goal of SQL Backup 7 is to make it so easy to schedule regular restores of your backups that you have no excuse not to. Why schedule a restore? Because your data is not as good as your last backup. It's only as good as your last successful restore. If you're not checking your backups by restoring them and running an integrity check on the database, you're only doing half the job. It seems that most DBAs know that this is best practice, but it can be tricky and time-consuming to set up, so it's one of those tasks that can get forgotten in the midst all the other demands on their time. Sometimes, they're just too busy firefighting. But if it was simple to do? That was our inspiration for SQL Backup 7. So it was heartening to read Brent Ozar's blog post the other day about World Backup Day. To be honest, I'd never heard of World Backup Day (Talk Like a Pirate Day, yes, but not this one); however, its emphasis on not just backing up your data but checking the validity of those backups was exactly the same message we had in mind when building SQL Backup 7. It's printed on a piece of A3 above our planning board - "Make backup verification so easy to do that no DBA has an excuse for not doing it" It's the missing piece that completes the puzzle. Simple idea, great concept, useful feature, but, as it turned out, far from straightforward to implement. The problem is the future. As Marty McFly discovered over the course of three movies, the future is uncertain and hard to predict - so when you are scheduling a restore to take place an hour, day, week or month after the backup, there are all kinds of questions that you wouldn't normally have to consider. Where will this backup live? Will it even exist at the time? Will it be split into multiple files? What will the file names be? Will it be encrypted? What files should it be restored to? SQL Backup needs to know what to expect at the time the restore job is actually run. Of course, a DBA will know the answer to all these questions, but to deliver the whole point of version 7, we wanted to make it easy for them to input that information into SQL Backup. We think we've done that. When you create your scheduled backup job, there is now an option to create a "reminder" to follow it up with a scheduled restore to verify the resulting backups. Actually, it's much more than a reminder, as it stores all the relevant data so you can click it and pre-populate the wizard with all the right settings to set up your verification restores. Simple. But, what do you think? We'd love you to try it. Post by Brian Harris

    Read the article

  • Anyone code at a treadmill desk? [closed]

    - by Sequenzia
    have been thinking about getting a treadmill desk for awhile now but I just don't know if it is possible to code at one. I can see doing a normal computer job while walking very slow but I just don't know if you can write code do it. Like a lot of people I could stand to lose weight and I am just not in shape anymore. I sit at my computer for at least 12 hours a day and then I am on my laptop for a few more hours. I need to do something to help my health. I also have been seeing a lot of reports about the long term health issues related to desk jobs. Like this. Before I drop a few hundred dollars on a new desk I am wondering if anyone has tried a treadmill desk and if so which one?

    Read the article

  • Positive reinforcements @ work [closed]

    - by nurne
    I found out that what fuels me to do well at work are positive reinforcements From bosses, colleagues, and customers My current job at a startup is very demanding My boss doesn't have time to give positive reinforcements, and also i'm always behind schedule so maybe i don't deserve positive reinforcements On the other hand i don't get any negative reinforcements, so i guess that as long as this doesn't happen - what i'm doing is ok How is your relationship with bosses colleagues and customers @ work? Do you need positive reinforcements? Do you get them? How do you make them happen? Is there some kind of standard for developers? For hi-tech? Thanks

    Read the article

  • 12.04 server on home network

    - by dustin mantei
    I need advice to help me with a server install, I'm new to ubuntu and linux in general. I have 6 systems in my house, 5 being windows 7 and this laptop that I am typing on is Linux mint 13 Maya. Question is, my wife will not transition to anything. She is stuck in Gates-land. so, can I make a server with ubuntu 12.04 (burnt disc image last night) so that all systems in my home can access and my mother-in-law in another state can also access with a username? That would be awesome and it may convince the war dept (my wife) to change all the systems in my house to ubuntu/linux. Sorry this is so long winded, but all the questions I have seen on this forum don't answer it completely.

    Read the article

  • Ubuntu install problem: "USB disconnect address 2"

    - by explorex
    Hi, I am installing ubuntu 10.10 and i am having problem in "Who are you?" section. How long will it take to install. i have been waiting for more than 20 minutes. and still i am not being able to click forward button. in ready when you are ... console i am getting message curr DATE ubuntu kernel: [ 684.152061 ] usb 5-1:USB disconnect address 2 whats happening i have no idea, but live stick is working fine. I am installing for USB drive and i have internet turned off edit:: now i am getting message curr Date ubuntu: CRON[9268]: (root) CMD (start -q anacron || :) for past 15 minutes

    Read the article

  • Oracle VM Virtualbox 4.0 released !

    - by wim.coekaerts
    Another great day for the Virtualbox development team. As is custom, they churn out new features and enhancements at a record pace. You can find the changelog here and you can download your version of 4.0 here. Have at it. A bunch of changes, visually a new management console, a new method to install with a base install and the extention pack (for the add-on drivers and extra features), a number of bugfixes, multi-monitor support for Oracle Solaris and Linux, it's a long list. A great product with a great user base. Check out this survey!

    Read the article

  • How to connect to the internet reliably

    - by user12689
    I have a DSL internet connection. For a long time the internet functioned fine in Ubuntu using the default Network Manager. Recently however, my ISP was bought by another ISP, and the service name had to be changed in the DSL settings. In Windows, it connects without any issue whatsoever. But in Ubuntu, the first time I connect I get a "Apache: Not Found" page, the next time, I get to the previous ISP's login page, it continues like this for some time until I get it working. I have to try connecting it again and again. When I asked the customer service, they said that because of the takeover, they have switched servers. But they said, it should work properly in Windows, and they were clueless about Ubuntu. Is there anything I can do to solve this issue?

    Read the article

  • How best to take a users signature online? (UK law orientated) [closed]

    - by Ben Griffiths
    Not sure if this is the best place to ask, but I can't seem to find any of the other SE sites that would fit better (unless there's a law one?) I'm building an application that will replace an existing paper based form, and this form would normally be signed by the person filling it in. Looking around, it's hard to find a good definitive resource to explain what I can and cannot accept as far as a signature goes. It looks like some UK government online forms accept just your name typed into a box, but I've also heard you should back up with an email - so that process would be type name into a box along with providing an email address, send out an email, then make them click a link within the email to finally complete the verification. Involving email seems very long winded and leaves the system open to spam filters blocking emails, forgotten emails that just sit in inbox's etc. So, does anyone have any knowledge in this department? Personally, I'd love to just get them to type their name into a box and be done with it!

    Read the article

  • What are interesting research questions with regards to open source software development? [closed]

    - by Aron Lindberg
    Imagine you have funding for a team of social scientists to study open source software development for a number of years (long time in software development, I know, but a short time for scientific research). These scientists have competencies to investigate psychological and sociological aspects of open source software development (i.e. how coders think, feel, and behave, along with how communities work or do not work). They are also technically equipped to understand code and coding, have access to all sorts of statistical and machine learning techniques, however their focus is on social aspects of open source software development, not technical. For you, as an open source software developer, what would be the research questions that would be interesting for you to have answered by such a group of scientists?

    Read the article

  • Test Doubles : Do they go in "source packages" or "test packages"?

    - by sbrattla
    I've got a couple of data access objects (DefaultPersonServices.class, DefaultAddressServices.class) which is responsible for various CRUD operations in a database. A few different classes use these services, but as the services requires that a connection is established with a database I can't really use them in unit tests as they take too long. Thus, I'd like to create a test doubles for them and simply do FakePersonServices.class and FakeAddressService.class implementations which I can use throughout testing. Now, this is all good (I assume)...but my question relates to where I put the test doubles. Should I keep them along with the default implementations (aka "real" implementations) or should I keep them in a corresponding test package. The default implementations are found in Source Packages : com.company.data.services. Should I keep the test doubles here too, or should the test doubles rather be in Test Packages : com.company.data.services?

    Read the article

  • Language Design: Are languages like phyton and coffescript really more comprehendable?

    - by kittensatplay
    the "Verbally Readable !== Quicker Comprehension" arguement on http://ryanflorence.com/2011/case-against-coffeescript/ is really potent and interesting. i and im sure other would be very interested in evidence arguing against this. there's clear evidence for this and i believe it. ppl naturally think in images, not words, so we should be designing languages dissimilar to human language like english, french, whatever. being "readable" is quicker comprehension. most articles on wikipedia are not readable as they are long, boring, dry, sluggish, very very wordy, and because wikipedia documents a ton of info, is not especially helpful when compared to much more helpful sites with more practical, useful, and relevant info. but languages like phyton and coffescript are "verbally readable" in that they are closer to the english language syntax, and programming firstly and mainly in python, im not so sure this is really a good thing. the second interesting argument is that coffeescript is an intermediator so thereby another step between to ends, which may increase chances of bugs. while coffeescript has other practical benefits, this question is focused specifically on evidence showing support for the counter-case of language "readability"

    Read the article

  • Handle all authentication logic in database or code?

    - by Snuffleupagus
    We're starting a new(ish) project at work that has been handed off to me. A lot of the database sided stuff has been fleshed out, including some stored procedures. One of the stored procedures, for example, handles creation of a new user. All of the data is validated in the stored procedure (for example, password must be at least 8 characters long, must contain numbers, etc) and other things, such as hashing the password, is done in the database as well. Is it normal/right for everything to be handled in the stored procedure instead of the application itself? It's nice that any application can use the stored procedure and have the same validation, but the application should have a standard framework/API function that solves the same problem. I also feel like it takes away the data from the application and is going to be harder to maintain/add new features to.

    Read the article

< Previous Page | 194 195 196 197 198 199 200 201 202 203 204 205  | Next Page >