Search Results

Search found 9091 results on 364 pages for 'what were you thinking'.

Page 166/364 | < Previous Page | 162 163 164 165 166 167 168 169 170 171 172 173  | Next Page >

  • Keeping a domain model consistent with actual data

    - by fstuijt
    Recently domain driven design got my attention, and while thinking about how this approach could help us I came across the following problem. In DDD the common approach is to retrieve entities (or better, aggregate roots) from a repository which acts as a in-memory collection of these entities. After these entities have been retrieved, they can be updated or deleted by the user, however after retrieval they are essentially disconnected from the data source and one must actively inform the repository to update the data source and make is consistent again with our in-memory representation. What is the DDD approach to retrieving entities that should remain connected to the data source? For example, in our situation we retrieve a series of sensors that have a specific measurement during retrieval. Over time, these measurement values may change and our business logic in the domain model should respond to these changes properly. E.g., domain events may be raised if a sensor value exceeds a predefined threshold. However, using the repository approach, these sensor values are just snapshots, and are disconnected from the data source. Does any of you have an idea on how to solve this following the DDD approach?

    Read the article

  • Is learning C# as a first language a mistake?

    - by JuniorDeveloper1208
    I know there are similar questions on here, which I've read, but I recently read this post by Joel Spolsky: How can I teach a bright person, with no programming experience, how to program? And it got me thinking about my way of learning and whether it might actually be harmful in the long run. I've dabbled with various languages but C# is my first serious one, I've read "Head First C#" and created a few projects. But after reading the post above I've found it a bit disheartening that I may be going about it all wrong, obviously I respect Joel's opinion which is what has thrown me a bit. I've started reading "Code" as recommended in the reading list and I'm finding it pretty hard going, although enjoyable. I feel like it's taken the shine off of my "noobish hacking about" in Visual Studio. So now I'm unsure as to what path I should take? Should I take a step back and follow Joel's advice and start reading? I guess my main aim is just to become a good programmer, like everyone else, but I don't want to be going into bad practice by learning a .NET language when someone who's opinion I respect thinks that it is harmful. Thoughts?

    Read the article

  • Should I use a separate 'admin' user as my "root sudo" or grant sudo to my 'app' user?

    - by AJB
    I'm still wrapping my brain around the Ubuntu 'nullify root' user management philosophy (and Linux in general) and I'm wondering if I should 'replace' my root user with a user called 'admin' (which basically has all the powers of the root, when using sudo) and create another user called 'app' that will be the primary user for my app. Here's the context: I'll be running a LNMP stack on Ubuntu 12.04 Server LTS. There will be only one app running on the server. The 'app' user needs to have SUPER privileges for MySQL. PHP will need to be able to exec() shell commands. The 'app' user will need to be able to transfer files via SFTP. And I'm thinking this would be the best approach: nullify 'root' user create a user called 'admin' that will be a full sudoer of root, this will be the new "root" user of NGINX, PHP, and MySQL (and all system software) grant SUPER privileges to 'app' in MySQL Grant SFTP privileges to only the 'app' user. As I'm new to this, and the information I've found in researching it tends to be of a more general nature, I'm wondering if this is a solid approach, or if it's unorthodox in a way that would cause issues down the road. Thanks in advance for any help.

    Read the article

  • Suggestion for setting web application parameters

    - by user40730
    I'm creating a web application on GWT. I'm using MVP pattern with activities and places. I have a xml config file containing some parameters to be used by the application. Content of this xml file is sent to the client using HttpRequest; I'm using a singleton class to hold the information from the xml file. Right now, the application is getting the data when the user starts the application in the home page, that is working well. Now, since I'm using activities and places, a user can bookmark a page and starts the application in any other page (Place). And here comes the problem: Since I'm using some of the information from the xml file to set some ui widgets, I have to check if the xml config file was read and the application already has the parameters (I do this by checking the singleton class). But the xml file is read by using an HttpRequest, so I got errors 'cause the application needs some parameters to initialize some ui widgets, but these parameters aren't ready on time. I was thinking on using an synchronous request to fix the problem, but it seems complicated and not recommendable to do that. So, I'd like to hear some other suggestions. Thanks.

    Read the article

  • Do cross reference database tables have a place in domain driven design?

    - by Mike Cellini
    First some background. Let's say we have a system where a customer is placing an order in a web interface. The items that customer is ordering can priced in various ways. Sometimes including the cost of delivery and sometimes not at all. That pricing effectively depends on a variety of factors including the vendor's own pricing model, that vendor's individual contracts with customers as well as that vendor's contracts with its own suppliers. Let's assume that once a customer places an order for a particular item and chooses a contract if any, the method of delivery can be determined by variables on those contracts. Those delivery methods also live in their own table in the database and have various properties consumed downstream. It makes sense that a cross reference or lookup table would store that information. That table would be loaded into the domain and could then be used to apply the appropriate delivery method while processing the order. Does this make sense in the context of domain driven design? Or is my thinking too relational? Is this logic that should be built into it's own class/method (I mean beyond apply the cross reference table data)?

    Read the article

  • Is it appropriate to run a complex enterprise-system configuration and migration project in a similar way to a Scrum development project?

    - by AndyM
    I'm just starting out on the implementation of a large enterprise-wide system, which has complex requirements and many stakeholders. The company has been through high-level evaluation and tender process and determined to purchase a highly configurable "off-the-shelf" product rather than building an entirely bespoke system. The system will replace several existing systems and will require a significant amount of data migration. I'm thinking that the implementation of this system (which is expected to take over 2 years) could be run in a similar way to a Scrum software development project. With the first sprints targeted at building the minimal possible functionality needed (across all functional areas), and then iteratively deepening the level of functionality according the stakeholder feedback. I think this will de-risk the project and help ensure a balance of stakeholder needs within the available time. The user stories are still the same, it's just that to implement them we have work within the constraints of the pre-purchased system. When it comes to 'building stuff', instead of writing custom code the team will be configuring the off-the-shelf package, writing data conversion scripts and the like (and it should be a lot quicker!). Does this sound like a sensible approach? Does the Agile approach makes sense here?

    Read the article

  • Cookie access within a HTTP Class

    - by James Jeffery
    I have a HTTP class that has a Get, and Post, method. It's a simple class I created to encapsulate Post and Get requests so I don't have to repeat the get/post code throughout the application. In C#: class HTTP { private CookieContainer cookieJar; private String userAgent = "..."; public HTTP() { this.cookieJar = new CookieContainer(); } public String get(String url) { // Make get request. Return the JSON } public String post(String url, String postData) { // Make post request. Return the JSON } } I've made the CookieJar a property because I want to preserve the cookie values throughout the session. If the user is logged into Twitter with my application, each request I make (be it get or post) I want to use the cookies so they remain logged in. That's the basics of it anyway. But, I don't want to return a string in all instances. Sometimes I may want the cookie, or a header value, or something else from the request. Ideally I'd like to be able to do this in my code: Cookie cookie = http.get("http://google.com").cookie("g_user"); String g_user = cookie.value; or String source = http.get("http://google.com").body; My question - To do this, would I need to have a Get class, and a Post class, that are included within the HTTP class and are accessible via accessors? Within the Get and Post class I would then have the Cookie method, and the body property, and whatever else is needed. Should I also use an interface, or create a Request class and have Post and Get extend it so that common methods and properties are available to both classes? Or, am I thinking totally wrong?

    Read the article

  • I am not the most logically-organized person. Do I have any chance at being a good 'low-level' programmer?

    - by user217902
    Background: I am entering college next year. I really enjoy making stuff and solving logical problems, so I'm thinking of majoring in compsci and working in software development. I hope to have the kind of job where I can work with implementing / improving algorithms and data structures on a regular basis.. as opposed to, say, a job that's purely concerned with mashing different libraries together, or 'finding the right APIs for the job'. (Hence the word 'low-level' in the title. No, I don't wish to write assembly all day.) Thing is, I've never been the most logically-sharp person. Thus far I have only worked on hobby projects, but I find that I make the silliest of errors ever so often, and it can take me ages to find it. Like anywhere between three hours to a day to locate a simple segfault, off-by-one error, or other logical mistake. (Of course, I do other things in the meantime, like browsing SO, reddit, and the like..) It's not like I'm 'new' to programming either; I first tried C++ maybe five years ago. My question is: is this normal? Should a programmer with any talent solve it in less time? Having read Spolsky's Smart and gets things done, where he talks about the large variance in programming speed, am I near the bottom of the curve, and therefore destined to work in companies that cannot afford to hire quality programmers? I'd like to think that conceptually I'm okay -- I can grasp algorithms and concepts pretty well, I do fine in math and science, although I probably drop signs in my equations more often than the next guy. Still, grokking concepts makes me happy, and is the reason why I want to work with algorithms. I'm hoping to hear from those of you with real-world programming experience. TL;DR: I make many careless mistakes, should I not consider programming as a career?

    Read the article

  • Restoring backup with Deja Dup from external HD

    - by widgg
    So, here's the problem that I have with restoring. First, I backed up like everything. So I had place on the external HD for only one copy. I had to reinstall everything, but I didn't worry because I had a backup. But now, restoring doesn't work and it starts to be annoying. So, I right click "Restore missing files...". Then I have the popup window from Deja Dup asking where is the backup. So, I select the external HD and either put nothing in "folder" or just put ".". Thinking that it should be the base to look for backups. In both cases, after a while scanning, I get: The Volume "Filesystem root" has only 139.7mb disk space remaining. But my partition "/home" has 799.6Gb free. Also, I just want to restore some files, I don't need all of them. On my external HD, there's a file called: duplicity-full.20120514T220834Z.manifest. A text file. In it, I can see that everything is partitions in files of 52mb. So, small files are in one archive while very large files are split in multiple one. But I can see the exact list of file that I have. So, I'm guessing that my backup is intact. What am I doing wrong ? Is possible that it failes, even to list all the files, because I don't have enough space left on my ExtHD ? Why can't duplicity used another location to do that ?

    Read the article

  • snapping an angle to the closest cardinal direction

    - by Josh E
    I'm developing a 2D sprite-based game, and I'm finding that I'm having trouble with making the sprites rotate correctly. In a nutshell, I've got spritesheets for each of 5 directions (the other 3 come from just flipping the sprite horizontally), and I need to clamp the velocity/rotation of the sprite to one of those directions. My sprite class has a pre-computed list of radians corresponding to the cardinal directions like this: protected readonly List<float> CardinalDirections = new List<float> { MathHelper.PiOver4, MathHelper.PiOver2, MathHelper.PiOver2 + MathHelper.PiOver4, MathHelper.Pi, -MathHelper.PiOver4, -MathHelper.PiOver2, -MathHelper.PiOver2 + -MathHelper.PiOver4, -MathHelper.Pi, }; Here's the positional update code: if (velocity == Vector2.Zero) return; var rot = ((float)Math.Atan2(velocity.Y, velocity.X)); TurretRotation = SnapPositionToGrid(rot); var snappedX = (float)Math.Cos(TurretRotation); var snappedY = (float)Math.Sin(TurretRotation); var rotVector = new Vector2(snappedX, snappedY); velocity *= rotVector; //...snip private float SnapPositionToGrid(float rotationToSnap) { if (rotationToSnap == 0) return 0.0f; var targetRotation = CardinalDirections.First(x => (x - rotationToSnap >= -0.01 && x - rotationToSnap <= 0.01)); return (float)Math.Round(targetRotation, 3); } What am I doing wrong here? I know that the SnapPositionToGrid method is far from what it needs to be - the .First(..) call is on purpose so that it throws on no match, but I have no idea how I would go about accomplishing this, and unfortunately, Google hasn't helped too much either. Am I thinking about this the wrong way, or is the answer staring at me in the face?

    Read the article

  • Is there a schematic overview of Ubuntu's architecture?

    - by joebuntu
    Hi there, as enthusiastic, advanced Linux learner, I'd love to get an overview about Linux' architecure/structure in general. You know, like "the big picture". I'm thinking of a large schematic graphic showing what is what, who is who, what system (e.g. X) comprises which subsystems (GDM/Gnome/Compiz) on the way from a to z, from boot to interactive desktop, including the most important background services (auth, network, cron, ...). Maybe a bit like this: http://www.flickr.com/photos/pgc/140859386/ but way more detailed. There's bootchart, which produces very comprehensive charts, but they again are too detailed and difficult to get the "big picture" from. Is there such a thing? Possibly not for the whole System, but maybe for single subsystems? I had trouble searching for this, because using search terms like "scheme" or "architecture" pointed to the wrong direction (a tool called "scheme" or CAD software for linux). I appreciate any links. If there's interest in those schematic overviews and links, maybe someone could turn this post into a wiki post? Cheers, joebuntu

    Read the article

  • Samba issue with sharing directories on NTFS/FAT32 (Mounted Drives) ???

    - by Microkernel
    Hi guys, I have some strange problems with Samba server. I am using samba Version 3.5.4 on Ubuntu 10.10. I have two windows-xp machines, one on VirtualBox on Ubuntu and another office laptop. Windows machine on VBox has no issues in accessing the shared folders, but the laptop is not able to access all the shared content. The issue faced on laptop is = Shared folders on Ext3 drives have no issues in accessing, but the contents shared on NTFS and FAT32 drives (mounted ones) are not accessible. When I try to open the shared folder, it asks for user name and password, but doesn't accept when I provide it. (even if I provide admin login details!!!). I changed workgroup value to the domain_name in office laptop, but still the problem persists... Here is the smdb.conf I am using... [global] workgroup = XXX.XXX.ORG server string = %h server (Samba, Ubuntu) map to guest = Bad User obey pam restrictions = Yes pam password change = Yes passwd program = /usr/bin/passwd %u passwd chat = *Enter\snew\s*\spassword:* %n\n *Retype\snew\s*\spassword:* %n\n *password\supdated\ssuccessfully* . unix password sync = Yes syslog = 0 log file = /var/log/samba/log.%m max log size = 1000 dns proxy = No usershare allow guests = Yes panic action = /usr/share/samba/panic-action %d guest ok = Yes [homes] comment = Home Directories [printers] comment = All Printers path = /var/spool/samba read only = No create mask = 0700 printable = Yes browseable = No [print$] comment = Samba server's CD-ROM path = /cdrom force user = nobody force group = nobody locking = No Workgroup Was defined as "HOMENET" before, changed it to domain name on the office laptop thinking it was the problem, but for no avail Thanks in advance Regards, Microkernel

    Read the article

  • Dual boot Win/7 and Ubuntu 12.04

    - by Brian
    So I've always wanted Ubuntu along with windows 7 and finally I went to try and make my computer that last night. (( I'll just skip the long story. )) I loaded Ubuntu from CD, Version 12.04 and it loaded fine and then I clicked the install icon. It asked me witch option I would like to proceed with, and I answered Ubuntu, in Windows 7. So it did everything rather quickly, and restarted itself; upon restarting itself it ejected the CD, and I thought everything was good to go. It brought me to the Option to either load Ubuntu or Windows, I was thinking to myself wow that was alot easier than I thought; Windows 7 Loads fine after it checked the HDD but when I go to load Ubuntu it brings me to the loading screen and then stays there for a long period time, finally moves on as if it was going to load into the regular dashboard, but loads into a dos looking thing. (( I'm sorry if I sound retarded explaining everything I'm not great with computers )) And at the very time it says something like installation failed. It also says it could not find a file or something like that. If you need me to go back and get the full message and put it up here I will. But if I put the CD in it loads fine. Thanks in advance for everyone that helps me solve this problem.

    Read the article

  • .NET app - Should we use SQL Server and duplicate some reference data from an external Oracle DB? Or use Oracle and have a DB link?

    - by Daventry
    We're looking to migrate some existing Excel/Access processes into a new system which will provide the users with a Silverlight frontend to run and view the reports instead of using MS Access. The initial idea was to have SQL Server 2008 as RDBMS. The problem is that we've got some static data such as country codes, counterparties, etc which live in an existing Oracle DB. Since we do not want to duplicate that data (if possible), we were thinking of having a DB link between SQL Server and Oracle, but our firm does not allow that. So the options are either duplicate the data or use Oracle as RDBMS - surprise, the firm does allow DB links between Oracle databases. The initial idea was also to use WCF RIA Services, Entity Framework, etc which we're not sure they play well with Oracle, that's why it was decided to go with SQL Server in the first place. Would you advise to go for Oracle so that we can just link the static data? Or use SQL Server 2008 and replicate it because it's "safer" to stay within the Microsoft land? To use or not to use Entity Framework and WCF RIA Services at all? Regards. UPDATE: Thanks everyone for your answers. Nothing is set in stone yet. We'll try to import the data instead of linking, as if the other DB goes down, our system can still carry on. We're likely to use SQL Server just because most developers are more experienced with it. Even if we used RIA Services, we can swap out the Data Access Layer and use other frameworks such those mentioned below.

    Read the article

  • What are some ways to separate game logic from animations and the draw loop?

    - by TMV
    I have only previously made flash games, using MovieClips and such to separate out my animations from my game logic. Now I am getting into trying my hand at making a game for Android, but the game programming theory around separating these things still confuses me. I come from a background of developing non game web applications so I am versed in more MVC like patterns and am stuck in that mindset as I approach game programming. I want to do things like abstract my game by having, for example, a game board class that contains the data for a grid of tiles with instances of a tile class that each contain properties. I can give my draw loop access to this and have it draw the game board based on the properties of each tile on the game board, but I don't understand where exactly animation should go. As far as I can tell, animation sort of sits between the abstracted game logic (model) and the draw loop (view). With my MVC mindset, it's frustrating trying to decide where animation is actually supposed to go. It would have quite a bit of data associated with it like a model, but seemingly needs to be very closely coupled with the draw loop in order to have things like frame independent animation. How can I break out of this mindset and start thinking about patterns that make more sense for games?

    Read the article

  • Using a random string to authenticate HMAC?

    - by mrwooster
    I am designing a simple webservice and want to use HMAC for authentication to the service. For the purpose of this question we have: a web service at example.com a secret key shared between a user and the server [K] a consumer ID which is known to the user and the server (but is not necessarily secret) [D] a message which we wish to send to the server [M] The standard HMAC implementation would involve using the secret key [K] and the message [M] to create the hash [H], but I am running into issues with this. The message [M] can be quite long and tends to be read from a file. I have found its very difficult to produce a correct hash consistently across multiple operating systems and programming languages because of hidden characters which make it into various file formats. This is of course bad implementation on the client side (100%), but I would like this webservice to be easily accessible and not have trouble with different file formats. I was thinking of an alternative, which would allow the use a short (5-10 char) random string [R] rather than the message for autentication, e.g. H = HMAC(K,R) The user then passes the random string to the server and the server checks the HMAC server side (using random string + shared secret). As far as I can see, this produces the following issues: There is no message integrity - this is ok message integrity is not important for this service A user could re-use the hash with a different message - I can see 2 ways around this Combine the random string with a timestamp so the hash is only valid for a set period of time Only allow each random string to be used once Since the client is in control of the random string, it is easier to look for collisions I should point out that the principle reason for authentication is to implement rate limiting on the API service. There is zero need for message integrity, and its not a big deal if someone can forge a single request (but it is if they can forge a very large number very quickly). I know that the correct answer is to make sure the message [M] is the same on all platforms/languages before hashing it. But, taking that out of the equation, is the above proposal an acceptable 2nd best?

    Read the article

  • Directing Multiple ccTLD's to 1 gTLD with a country specific subdirectory?

    - by Pascal Van Opzeeland
    We have multiple ccTLDomains and are thinking about how to best combine these into one. We want to do this to focus our link building efforts. We are running a website through which we offer a software-as-a-service. Therefore we could potentially sell to any country in the world. However, Germany is our most important market. We currently have a .com, .de, .nl. and .pl domain. All these domains have a high amount of unique content pages. What we are planning is to change everything to .com with language-based subdirectories, so .com/en/, .com/de/, etc. I have two questions concerning this issue: 1) How much of an advantage does a ccTLD have over a gTLD with country specific subdirectories in search rankings? So let’s say .de versus .com/de/? 2) How could we best redirect the visitors of our old ccTLD’s to our gTLD’s subdirectories? We would like to loose as few search engine rankings as possible. Thank you for your help.

    Read the article

  • Asus A8V overcurrent

    - by user139710
    This is not as much as a question as it is a note to those out there that upgrade their motherboards with better processors and the like. Here's my story. Recently I upgraded my processor. System specifications: • Asus A8V Deluxe • 4GB RAM • ATI Radeon 3870 AGP graphics card (I believe that's it) Anyway, I decided to put a dual core Opteron 180 in this rig, but the problem was that I needed to update the BIOS to V-1017, and not knowing the consequences, I went up to the Asus site and got the newest, the latest and the greatest, 1018.002 thinking that it was the best for this board, however it wasn't. I used the Asus EXFlash, which makes life a lot easier, flashed the BIOS and all of a sudden I start getting this message: USB overcurrent protection, system shutting down in 15 seconds to protect your system. WELL SHIT... This is a new one on me... I read the blogs, all the posts on this thing, and did all that everyone else did to correct the problem, but nothing helped. So i decided to start from square one, went back to Asus and looked at the BIOS download... OMG... IT WAS A BETA. So, I downloaded the update that was suggested 1017< and installed it and wouldn't you know, it took care of the problem, no more USB overcurrent protection, no more crashing. I write this today to let you all know about this, just in case you have an issue such as this. Well there you all go. Fly safe and eat your vegetables.

    Read the article

  • Is it worth replacing mouse by standalone trackpad for heavy code-editing? [on hold]

    - by heltonbiker
    I recently got more interested in improving my tools, workspace and worflow. The first sting came with a sore finger due to a crappy keyboard, and then after some research I fell in love with the "mechanical keyboard is what you need" doctrine, bought one (cherry MX Brown if you're curious), and am very happy with the results. Currently I am replacing my previous text editor (Geany) with Sublime Text 3, and am also very happy and feeling much more powerful and professional :) Well, but while I re-read all the ancient debates about VIM vs whatever-else, the following excerpt from a blog post got me thinking again about the mouse vs keyboard, and the "moving around from the very home row" (in VIM) versus gesturing away with the tiny and unstable mouse cursor: Reaching for a mouse may indeed slow you down, but developers are commonly on machines where the trackpad is a micro-hand movement away. Most novice programmers can click on a character on screen faster than an expert Vimmer can type 20jFp; or LkEEE or /word or any other nasty way Vimmers have to use. The point of a mouse is to make arbitrary on screen jumps efficient, and it’s very good at doing that. Don’t you ever think you can beat a mouse. Well, although there is some bitterness in this statement, it makes a lot of sense, and EVEN MORE if you consider your direct input to be a TRACKPAD conveniently placed in front of your spacebar (which oddly is where I like to put my mouse, rotated 90° ccw, due to a serious tendonitis in my right shoulder, already healed, but you knod...). So, the question is: Has anyone replaced mouse by a standalone trackpad, to work in code editing in a desktop machine (that is, with a sandalone keyboard)? Was it worth the change?

    Read the article

  • Testing a codebase with sequential cohesion

    - by iveqy
    I've this really simple program written in C with ncurses that's basically a front-end to sqlite3. I would like to implement TDD to continue the development and have found a nice C unit framework for this. However I'm totally stuck on how to implement it. Take this case for example: A user types a letter 'l' that is captured by ncurses getch(), and then an sqlite3 query is run that for every row calls a callback function. This callback function prints stuff to the screen via ncurses. So the obvious way to fully test this is to simulate a keyboard and a terminal and make sure that the output is the expected. However this sounds too complicated. I was thinking about adding an abstraction layer between the database and the UI so that the callback function will populate a list of entries and that list will later be printed. In that case I would be able to check if that list contains the expected values. However, why would I struggle with a data structure and lists in my program when sqlite3 already does this? For example, if the user wants to see the list sorted in some other way, it would be expensive to throw away the list and repopulate it. I would need to sort the list, but why should I implement sorting when sqlite3 already has that? Using my orginal design I could just do an other query sorted differently. Previously I've only done TDD with command line applications, and there it's really easy to just compare the output with what I'm expected. An other way would be to add CLI interface to the program and wrap a test program around the CLI to test everything. (The way git.git does with it's test-framework). So the question is, how to add testing to a tightly integrated database/UI.

    Read the article

  • License validation and calling home

    - by VitalyB
    I am developing an application that, when bought, can be activated using a license. Currently I am doing offline validation which is a bit troubling to me. I am aware there is nothing to do against cracks (i.e modified binaries), however, I am thinking to trying to discourage license-key pirating. Here is my current plan: When the user activates the software and after offline validation is successful, it tries to call home and validate the license. If home approves of the license or if home is unreachable, or if the user is offline, the license gets approved. If home is reached and tells the license is invalid, validation fails. Licensed application calls home the same way every time during startup (in background). If license is revoked (i.e pirated license or generated via keygen), the license get deactivated. This should help with piracy of licenses - An invalid license will be disabled and a valid license that was pirated can be revoked (and its legal owner supplied with new license). Pirate-users will be forced to use cracked version which are usually version specific and harder to reach. While it generally sounds good to me, I have some concerns: Users tend to not like home-calling and online validation. Would that kind of validation bother you? Even though in case of offline/failure the application stays licensed? It is clear that the whole scheme can be thwarted by going offline/firewall/etc. I think that the bother to do one of these is great enough to discourage casual license sharing, but I am not sure. As it goes in general with licensing and DRM variations, I am not sure the time I spend on that kind of protection isn't better spent by improving my product. I'd appreciate your input and thoughts. Thanks!

    Read the article

  • Why does Android make good coding so difficult?

    - by metacircle
    my daily work is writing tools in C#/WPF. After over more than 1 year on the job now, I came to love MVVM, IoC Containers, XAML (and more). It's pure fun to write code, since simple, maintainable and extendable code just comes naturally when you follow a few basic patterns. In my free time I really want to write some apps, mainly for my own personal use. I want to write apps for fun and not to make money or anything, that being said, paying an annual fee to be allowed to use my own apps on my own device is a total no-go for me. So I am not able to code for Windows Phone and am also not able to use Xamarin on Android (which is sad since Visual Studio + Resharper is programmers heaven). So I am stuck with Android "classic" Java development. Everytime I sit down at home to create an app, or improve some of the code I have already written I get annoyed very quick because getting good, decoupled code is just so hard to accomplish. It feels like everything you have to do in Android to create a good architecture is a workaround instead of being the way things are meant to be. Writing the UI in xml is fine, but everything else is one big code mess. Even all the tutorials do all their coding in the code behind. For 'hello world' this is fine, but for anything bigger this gets messy very very quick. This is where the fun for me ends. It's just no fun anymore because I just spend 90% of my time refactoring and thinking of workarounds how to make my code more maintainable with all the restrictions Android puts on me. Am I missing a crucial part or is this just the way Android is meant to be? Do you have any suggestions how to learn 'the fun way' of Android programming.

    Read the article

  • shrink ext4 partition

    - by user276851
    My question is similar to Move ext4 partition, but the challenge I couldn't figure out is how to shrink a partition from the start. So suppose originally the partition (with raid) is like this. (************** /dev/md127 ***************) After resizing, I want to achieve like this. (*** unallocated ***)(**** /dev/md127 ****) Note, I cannot use gparted, and parted does not support ext4. The commands I have tried so far, % resize2fs -p /dev/md127 1676G # <== This is good. % lvreduce -L 1676G /dev/md127 Path required for Logical Volume "md127" Please provide a volume group name Run `lvreduce --help' for more information. Failed here, I guess it may be because the underlying partition is primary and the lvreduce only works on logical? Anyway, no idea. Then after that, I am thinking to create another partition right after this one, copy the data to that partition, and remove this one, like. 1. (************** /dev/md127 ***************) 2. (**** /dev/md127 ****)(*** new partition **) 3. (*** unallocated ****)(**** /dev/md127 ****) Thanks for the help.

    Read the article

  • Finding the best practice for a game simulating tool

    - by Tougheart
    I'm studying Java right now, and I'm thinking of this tool as my practice project. The game is "League of Legends" in case anyone knows it, I'm not actually simulating the game as in simulating game play, I'm just trying to create a tool that can compare different champions to each other based on their own abilities and items bought inside the game. The game basics are: Every player has a champion in a team of 5 players playing against another team. Each champion has a different set of abilities (usually 4) that s/he uses to do damage to opposing champions. Each champion gets stronger by buying different items, increasing the attack it deals or decreasing the damage received. What I want to do is to create a tool to be used outside the game enabling players to try out different builds for their champions and compare the figures against other champions they usually fight against. The goal is to enable players get a deeper understanding of the different item combinations (builds) that can be used during the games, instead of trying them out in real games which can be somehow very time consuming. What I'm stuck at is the best practice I should follow to make this possible using Java, I can't figure out which classes should inherit from which, should I make champions and items specs in the code or extracted from other files, specially that I'm talking about hundreds of items and champions to use in that tool. I'm self studying Java, and I don't have much practice at it, so I would really appreciate any broad guidelines regarding this, and sorry if my question doesn't fit here, I tried to follow the rules. English isn't my native language, so I'm really sorry if I wasn't clear enough, I would be more than happy to explain anything that's not understood.

    Read the article

  • Making a class pseudo-immutable by setting a flag

    - by scott_fakename
    I have a java project that involves building some pretty complex objects. There are quite a lot (dozens) of different ones and some of them have a HUGE number of parameters. They also need to be immutable. So I was thinking the builder pattern would work, but it ends up require a lot of boilerplate. Another potential solution I thought of was to make a mutable class, but give it a "frozen" flag, a-la ruby. Here is a simple example: public class EqualRule extends Rule { private boolean frozen; private int target; public EqualRule() { frozen = false; } public void setTarget(int i) { if (frozen) throw new IllegalStateException( "Can't change frozen rule."); target = i; } public int getTarget() { return target; } public void freeze() { frozen = true; } @Override public boolean checkRule(int i) { return (target == i); } } and "Rule" is just an abstract class that has an abstract "checkRule" method. This cuts way down on the number of objects I need to write, while also giving me an object that becomes immutable for all intents and purposes. This kind of act like the object was its own Builder... But not quite. I'm not too excited, however, about having an immutable being disguised as a bean however. So I had two questions: 1. Before I go too far down this path, are there any huge problems that anyone sees right off the bat? For what it's worth, it is planned that this behavior will be well documented... 2. If so, is there a better solution? Thanks

    Read the article

< Previous Page | 162 163 164 165 166 167 168 169 170 171 172 173  | Next Page >