Search Results

Search found 11778 results on 472 pages for 'mark ms smith'.

Page 112/472 | < Previous Page | 108 109 110 111 112 113 114 115 116 117 118 119  | Next Page >

  • What can you do to decrease the number of live issues with applications?

    - by User Smith
    First off I have seen this post which is slightly similar to my question. : What can you do to decrease the number of deployment bugs of a live website? Let me layout the situation for you. The team of programmers that I belong to have metrics associated with our code. Over the last several months our errors in our live system have increased by a large amount. We require that our updates to applications be tested by at least one other programmer prior to going live. I personally am completely against this as I think that applications should be tested by end users as end users are much better testers than programmers, I am not against programmers testing, obviously programmers need to test code, but they are most of the times too close to the code. The reason I specify that I think end users should test in our scenario is due to the fact that we don't have business analysts, we just have programmers. I come from a background where BAs took care of all the testing once programmers checked off it was ready to go live. We do have a staging environment in place that is a clone of the live environment that we use to ensure that we don't have issues between development and live environments this does catch some bugs. We don't do end user testing really at all, I should say we don't really have anyone testing our code except programmers, which I think gets us into this mess (Ideally, we would have BAs or QA or professional testers test). We don't have a QA team or anything of that nature. We don't have test cases for our projects that are fully laid out. Ok, I am just a peon programmer at the bottom of the rung, but I am probably more tired of these issues than the managers complaining about them. So, I don't have the ability to tell them you are doing it all wrong.....I have tried gentle pushes in the correct direction. Any advice or suggestions on how to alleviate this issue is greatly appreciated. Thanks.

    Read the article

  • Runtime analysis

    - by Joe Smith
    can someone please help me with the analysis of the following function (for inputs of size n). The part that confuses me the most is the inner for loop. def prefix_sums(L): # Total cost = ? pSum = [] #cost = 1 for a in range(len(L)+1): # range + body of function = (n+1) + (n+1)*(body) ? s = 0 #cost = 1 for b in range(a): # cost = ? s = s + L[b] #cost = operation + accessing list = 2 pSum.append(s) #cost = 1 return pSum #cost = 1 What I need to do is figure out the cost of each statement.

    Read the article

  • NDC Oslo

    - by Alan Smith
    Originally posted on: http://geekswithblogs.net/asmith/archive/2013/06/14/153136.aspx2013 has been a hectic year for conference presentations so far, NDC in Oslo has been the 6th conference I have attended, and my session there was my 11th conference presentation this year. I have been meaning to make the short trip over from Stockholm to NDC for a few years, and this was the first time I made it. I have heard a lot of great things about the event, and was impressed with the location, the sessions, and most of all the atmosphere around the event boots and during the party on Thursday evening. The session I was delivering was my “Grid Computing with 256 Windows Azure Worker Roles & Kinect” demo, which I have delivered at many events over the past 12 months. The demo went fine. I’m always a little nervous when I try to scale out the application to 256 worker roles, it almost always works well and the application will scale in minutes, but very occasionally there can be a longer delay due to the provisioning process in the Windows Azure data centers. This would not be an issue for many scenarios, but when standing on stage in front of a room full of developers you really want things to run smoothly. A number of people have suggested that I should pre-provision an environment so that it is guaranteed to be there when I run the demo during a session. For me the aim has always been to show the rapid scalability on cloud-based platforms live on stage. Pre-provisioning an environment may make for a more reliable demo but to me that would be cheating, and not half as much fun!

    Read the article

  • A Generic RIDC Test Program

    - by Kevin Smith
    Many times I have found it useful to use a java program that communicates with WebCenter Content (WCC) using RIDC for testing. I might not have access to the web GUI or need to test a service running as a specific user. In the past I had created a number of "one off" programs that submitted specific services, e.g GET_SEARCH_RESULTS, DOCINFO, etc. Recently I decided to create a generic RIDC test program that could submit any service with the desired parameters based on a configuration file. The programs gets the following information from the configuration file: WCC connection information (host, port) User to use to run service Service to run Any parameters for the service The program will make a connection to the WCC server, send the service request, and print the results of the service call using the getResponseAsString() method. Here is a sample configuration file: ridc.host=localhostridc.port=4444ridc.user=sysadminridc.idcservice=GET_SEARCH_RESULTSidcservice.QueryText=dDocType <matches> `Document`idcservice.SortField=dDocNameidcservice.SortDesc=ASC There is a readme file included in the zip with instructions for how to configure and run the program. The program takes one command line argument, the configuration file name. The configuration file name is optional and defaults to config.properties. If you have any suggestions for improvements let me know. Right now it only submits a single service call each time you run it. One enhancement I have already thought about would be to allow you to specify multiple services to tun in the configuration file. You can do that with the current program by having multiple configuration files and running the program multiple times, each with a different configuration file. You can download the program here.

    Read the article

  • What can be done to decrease the number of live issues with applications?

    - by User Smith
    First off I have seen this post which is slightly similar to my question. : What can you do to decrease the number of deployment bugs of a live website? Let me layout the situation for you. The team of programmers that I belong to have metrics associated with our code. Over the last several months our errors in our live system have increased by a large amount. We require that our updates to applications be tested by at least one other programmer prior to going live. I personally am completely against this as I think that applications should be tested by end users as end users are much better testers than programmers, I am not against programmers testing, obviously programmers need to test code, but they are most of the times too close to the code. The reason I specify that I think end users should test in our scenario is due to the fact that we don't have business analysts, we just have programmers. I come from a background where BAs took care of all the testing once programmers checked off it was ready to go live. We do have a staging environment in place that is a clone of the live environment that we use to ensure that we don't have issues between development and live environments this does catch some bugs. We don't do end user testing really at all, I should say we don't really have anyone testing our code except programmers, which I think gets us into this mess (Ideally, we would have BAs or QA or professional testers test). We don't have a QA team or anything of that nature. We don't have test cases for our projects that are fully laid out. Ok, I am just a peon programmer at the bottom of the rung, but I am probably more tired of these issues than the managers complaining about them. So, I don't have the ability to tell them you are doing it all wrong.....I have tried gentle pushes in the correct direction. Any advice or suggestions on how to alleviate this issue ?

    Read the article

  • One of my most frequently used commands

    - by Kevin Smith
    On a Linux or UNIX server this is one of my most frequently used commands. find . -name "*.htm" -exec grep -iH "alter session" {} \; It is an easy way to find a string you know is in a group of files, but don't know or can't remember which file it is in. For the example above, I knew that WebCenter Content sends a bunch of alter session commands to the database when it opens a new database connection. I wanted to find where these were defined and what all the alter session commands were. So, I ran these commands: cd /opt/oracle/middleware/Oracle_ECM1/ucm/idc/resources/core find . -name "*.htm" -exec grep -iH "alter session" {} \; And the results were: ./tables/query.htm: ALTER SESSION SET optimizer_mode = ?./tables/query.htm: ALTER SESSION SET NLS_LENGTH_SEMANTICS = ?./tables/query.htm: ALTER SESSION SET NLS_SORT = ?./tables/query.htm: ALTER SESSION SET NLS_COMP = ?./tables/query.htm: ALTER SESSION SET CURSOR_SHARING = ?./tables/query.htm: ALTER SESSION SET EVENTS '30579 trace name context forever, level 2'./tables/query.htm: ALTER SESSION SET NLS_DATE_FORMAT = ?./tables/query.htm: alter session set events '30579 trace name context forever, level 2' I could then go edit the query.htm file and find the include that contained all the ALTER SESSION commands.

    Read the article

  • Be aware of the difference between CURRENT_DATE and SYSDATE

    - by Kevin Smith
    I was running some queries in SQL Developer against the WebCenter Content (WCC) schema that included date fields such as dInDate. I was comparing the dates against CURRENT_DATE. I was not getting the expected results. I did some googlng and didn’t find a solution, but I did run across a reference to SYSDATE. I tried SYSDATE in my queries and got the expected results. I did a TO_CHAR on the two date fields and found they returned different times. CURRENT_DATE returned the time from my laptop which was  in the EDT time zone. SYSDATE returned the time from the database server which happened to be in the PDT time zone. I guess if both the database server and my laptop were in the same time zone I would not have seen any problem. Here is the query I ran to display the two fields. select to_char(current_date,'DD-MON-YY HH:MI:SS'), to_char(sysdate,'DD-MON-YY HH:MI:SS') from dual; As you can see from the screen shot from SQL Developer they definitely returned different times. I’m sure there is some command or setting you can use to prevent this problem, but for me the take away is to use SYSDATE in your queries when you want to do any date comparison.

    Read the article

  • Securing credentials passed to web service

    - by Greg Smith
    I'm attempting to design a single sign on system for use in a distributed architecture. Specifically, I must provide a way for a client website (that is, a website on a different domain/server/network) to allow users to register accounts on my central system. So, when the user takes an action on a client website, and that action is deemed to require an account, the client will produce a page (on their site/domain) where the user can register for a new account by providing an email and password. The client must then send this information to a web service, which will register the account and return some session token type value. The client will need to hash the password before sending it across the wire, and the webservice will require https, but this doesn't feel like it's safe enough and I need some advice on how I can implement this in the most secure way possible. A few other bits of relevant information: Ideally we'd prefer not to share any code with the client We've considered just redirecting the user to a secure page on the same server as the webservice, but this is likely to be rejected for non-technical reasons. We almost certainaly need to salt the password before hashing and passing it over, but that requires the client to either a) generate the salt and communicate it to us, or b) come and ask us for the salt - both feel dirty. Any help or advice is most appreciated.

    Read the article

  • Agilist, Heal Thyself!

    - by Dylan Smith
    I’ve been meaning to blog about a great experience I had earlier in the year at Prairie Dev Con Calgary.  Myself and Steve Rogalsky did a session that we called “Agilist, Heal Thyself!”.  We used a format that was new to me, but that Steve had seen used at another conference.  What we did was start by asking the audience to give us a list of challenges they had had when adopting agile.  We wrote them all down, then had everybody vote on the most interesting ones.  Then we split into two groups, and each group was assigned one of the agile challenges.  We had 20 minutes to discuss the challenge, and suggest solutions or approaches to improve things.  At the end of the 20 minutes, each of the groups gave a brief summary of their discussion and learning's, then we mixed up the groups and repeated with another 2 challenges. The 2 groups I was part of had some really interesting discussions, and suggestions: Unfinished Stories at the end of Sprints The first agile challenge we tackled, was something that every single Scrum team I have worked with has struggled with.  What happens when you get to the end of a Sprint, and there are some stories that are only partially completed.  The team in question was getting very de-moralized as they felt that every Sprint was a failure as they never had a set of fully completed stories. How do you avoid this? and/or what do you do when it happens? There were 2 pieces of advice that were well received: 1. Try to bring stories to completion before starting new ones.  This is advice I give all my Scrum teams.  If you have a 3-week sprint, what happens all too often is you get to the end of week 2, and a lot of stories are almost done; but almost none are completely done.  This is a Bad Thing.  I encourage the teams I work with to only start a new story as a very last resort.  If you finish your task look at the stories in progress and see if there’s anything you can do to help before moving onto a new story.  In the daily standup, put a focus on seeing what stories got completed yesterday, if a few days go by with none getting completed, be sure this fact is visible to the team and do something about it.  Something I’ve been doing recently is introducing WIP (Work In Progress) limits while using Scrum.  My current team has 2-week sprints, and we usually have about a dozen or stories in a sprint.  We instituted a WIP limit of 4 stories.  If 4 stories have been started but not finished then nobody is allowed to start new stories.  This made it obvious very quickly that our QA tasks were our bottleneck (we have 4 devs, but only 1.5 testers).  The WIP limit forced the developers to start to pickup QA tasks before moving onto the next dev tasks, and we ended our sprints with many more stories completely finished than we did before introducing WIP limits. 2. Rather than using time-boxed sprints, why not just do away with them altogether and go to a continuous flow type approach like KanBan.  Limit WIP to keep things under control, but don’t have a fixed time box at the end of which all tasks are supposed to be done.  This eliminates the problem almost entirely.  At some points in the project (releases) you need to be able to burn down all the half finished stories to get a stable release build, but this probably occurs less often than every sprint, and there are alternative approaches to achieve it using branching strategies rather than forcing your team to try to get to Zero WIP every 2-weeks (e.g. when you are ready for a release, create a new branch for any new stories, but finish all existing stories in the current branch and release it). Trying to Introduce Agile into a team with previous Bad Agile Experiences One of the agile adoption challenges somebody described, was he was in a leadership role on a team he had recently joined – lets call him Dave.  This team was currently very waterfall in their ALM process, but they were about to start on a new green-field project.  Dave wanted to use this new project as an opportunity to do things the “right way”, using an Agile methodology like Scrum, adopting TDD, automated builds, proper branching strategies, etc.  The problem he was facing is everybody else on the team had previously gone through an “Agile Adoption” that was a horrible failure.  Dave blamed this failure on the consultant brought in previously to lead this agile transition, but regardless of the reason, the team had very negative feelings towards agile, and was very resistant to trying it out again.  Dave possibly had the authority to try to force the team to adopt Agile practices, but we all know that doesn’t work very well.  What was Dave to do? Ultimately, the best advice was to question *why* did Dave want to adopt all these various practices. Rather than trying to convince his team that these were the “right way” to run a dev project, and trying to do a Big Bang approach to introducing change.  He would be better served by identifying problems the team currently faces, have a discussion with the team to get everybody to agree that specific problems existed, then have an open discussion about ways to address those problems.  This way Dave could incrementally introduce agile practices, and he doesn’t even need to identify them as “agile” practices if he doesn’t want to.  For example, when we discussed with Dave, he said probably the teams biggest problem was long periods without feedback from users, then finding out too late that the software is not going to meet their needs.  Rather than Dave jumping right to introducing Scrum and all it entails, it would be easier to get buy-in from team if he framed it as a discussion of existing problems, and brainstorming possible solutions.  And possibly most importantly, don’t try to do massive changes all at once with a team that has not bought-into those changes.  Taking an incremental approach has a greater chance of success. I see something similar in my day job all the time too.  Clients who for one reason or another claim to not be fans of agile (or not ready for agile yet).  But then they go on to ask me to help them get shorter feedback cycles, quicker delivery cycles, iterative development processes, etc.  It’s kind of funny at times, sometimes you just need to phrase the suggestions in terms they are using and avoid the word “agile”. PS – I haven’t blogged all that much over the past couple of years, but in an attempt to motivate myself, a few of us have accepted a blogger challenge.  There’s 6 of us who have all put some money into a pool, and the agreement is that we each need to blog at least once every 2-weeks.  The first 2-week period that we miss we’re eliminated.  Last person standing gets the money.  So expect at least one blog post every couple of weeks for the near future (I hope!).  And check out the blogs of the other 5 people in this blogger challenge: Steve Rogalsky: http://winnipegagilist.blogspot.ca Aaron Kowall: http://www.geekswithblogs.net/caffeinatedgeek Tyler Doerkson: http://blog.tylerdoerksen.com David Alpert: http://www.spinthemoose.com Dave White: http://www.agileramblings.com (note: site not available yet.  should be shortly or he owes me some money!)

    Read the article

  • What is your most preferred method of site pagination?

    - by John Smith
    There seem to be quite a few implementations of this feature. Some sites like like Stackexchange have it laid out like this: [1][2][3][4][5] ... [954][Next] Other sites like game forums may have something like this: [1][2][3] ... [10] ... [50] ... [500] ... [954][Next] Some sites like webcomics (XKCD comes to mind) have it laid out like this: [Last][Prev][Random][Next][First] Reddit has a very simple pagination with only: [Prev][Next] Sites like Stackexchange and Google also allow you to change how many results you want per page. Personally, I have never used this feature. Is it even worth including or does it just further confuse the design with needless features? Personally, I have only ever seen the need for the webcomic style (without the random). If I need to go to a specific page (which is very, very rare) then I can just edit the address bar. Is it good design to make something more complex for rare occasions where it might make save the user some time? Is having to edit the address bar to navigate the site effectively in some circumstances bad design?

    Read the article

  • If I were to claim I knew C++, what libraries would you expect me to know?

    - by Peter Smith
    I'm unsure as to the definition of knowing a programming language, so I'm picking C++ as an example. How much does it take to someone to be qualified as knowing C++? Should they just know the basic syntax? Template and generic-programming? Compiler flags and their purposes (Wall, the difference between O1, O2 and O3)? STL? Garbage collection strategies? Boost? Common libraries like zlib, curl, and libxml2?

    Read the article

  • When to skip solving the general problem and settling for the specific problem?

    - by Peter Smith
    I've been working hard on trying to develop a general solution to my problem, but I cannot seem to formulate a proper algorithm for it, at least one that doesn't take a ton of inaccurate grunt work building a lookup table. I have a solution already for the specific requirement, but it requires the software's configuration to be changed every time the software is loaded with a different geographic area's datasets. So is it better to be finished and move on for now, or to keep attempting to solve the general problem knowing that the specific problems will keep popping up?

    Read the article

  • Sweden Azure Group with Michele Laroux Bustamente &amp; Maartin Balliauw Thursday 22nd May

    - by Alan Smith
    Originally posted on: http://geekswithblogs.net/asmith/archive/2014/05/19/156418.aspxSweden Azure Group (SWAG) has the privilege of welcoming Michele Laroux Bustamente and Maartin Balliauw to present sessions at our meeting this Thursday. Michele and Maartin are two of the world’s leading experts in Cloud Computing and Azure, and will be taking time out from their busy schedules to share their ideas with us, and answer any questions. Knowit Stockholm are kindly hosting the event at their offices, and providing food and refreshments. It should be a great evening. You can register for the event here. Azure Q & A - Michele Leroux Bustamante In this interactive Q & A session Michele Leroux Bustamante will be on hand to share her wealth of experience on Azure related issues. If you are new to Azure and wanting some tips to get started, or an experienced developer needing to negotiate the legal and political protocols related to Cloud Computing Michele will have been there, done that, and be willing to share her experiences. This session will be entirely driven by that attendees, so please come prepared with questions. Reducing latency on the web with the Windows Azure CDN – Maarten Balliauw Serving up content on the Internet is something our web sites do daily. But are we doing this in the fastest way possible? How are users in faraway countries experiencing our apps? Why do we have three webservers serving the same content over and over again? In this session, we’ll explore the Windows Azure Content Delivery Network or CDN, a service which makes it easy to serve up blobs, videos and other content from servers close to our users. We’ll explore simple file serving as well as some more advanced, dynamic edge caching scenarios. Michele Leroux Bustamante Michele Leroux Bustamante is CIO at Solliance (solliance.net), cofounder of Snapboard (snapboard.com), and is recognized as a Microsoft Regional Director and MVP. Michele is a thought leader with over 20 years specializing in building scalable and secure end-to-end system design, identity and access management, and cloud computing technologies – for companies of all sizes. In recent years Michele has also helped launch several startup business ventures and has been a mentor to startups in several accelerator programs – providing both technical and business guidance. Michele shares her experiences through presentations and keynotes all over the world, and has been publishing regularly in technology journals. Maarten Balliauw Maarten Balliauw is a Technical Evangelist at JetBrains. His interests are all web: ASP.NET MVC, PHP and Windows Azure. He’s a Microsoft Most Valuable Professional (MVP) for Azure and an ASPInsider. He has published many articles in both PHP and .NET literature such as MSDN magazine and PHP architect. Maarten is a frequent speaker at various national and international events such as MIX (Las Vegas), TechDays, DPC, …

    Read the article

  • Link tags in iframe widget

    - by john Smith
    I have a rating community-site and I´m offering little iframe widgets with the average rating and some little other info. Does it make sense (for visibility, SEO) to add link tags to the head like: <link rel="alternate" type="application/rss+xml" title="RSS 2.0" href="rssfeed" /> <link rel="index" title="main-profile" href="main-profile"> To get a logical association of the widget to relating pages? How would you do this?

    Read the article

  • Problem installing from Ubuntu 12.04.1 LTS 32bit cd

    - by John Smith
    Older laptop currently running xp, only 128mb ram too. Is 128 just too small? But, 20+ gigs free hard drive and it's been defragmented. When I try to install Ubuntu from a CD I get the screen that says ubuntu and has the four red dots and then eventually goes blank and I just hear hard drive noises. Stays this way indefinitely (shut it off after half a day). Burned another cd, at slow writing speed too, and dl is from Ubuntu and get same result. Any help much appreciated!

    Read the article

  • Visual Studio 2012 Launch Winnipeg&ndash;Slides

    - by Dylan Smith
    The Winnipeg .Net User Group hosted a VS 2012 Launch Event at the Imax in Winnipeg on Thursday, Dec 6.  Doing presentations on the giant Imax screen is always fun, and I did the first 2 sessions on: End-To-End Application Lifecycle Management with TFS 2012 Improving Developer Productivity with Visual Studio 2012 Thanks to everybody that came out, and if anybody is interested my slide decks can be downloaded here: TFS 2012 Slides VS 2012 Slides Also the Virtual Machine that I used to do my demo’s can be downloaded from Brian Keller’s blog here: VS 2012 ALM Virtual Machine

    Read the article

  • circle - rectangle collision in 2D, most efficient way

    - by john smith
    Suppose I have a circle intersecting a rectangle, what is ideally the least cpu intensive way between the two? method A calculate rectangle boundaries loop through all points of the circle and, for each of those, check if inside the rect. method B calculate rectangle boundaries check where the center of the circle is, compared to the rectangle make 9 switch/case statements for the following positions: top, bottom, left, right top left, top right, bottom left, bottom right inside rectangle check only one distance using the circle's radius depending on where the circle happens t be. I know there are other ways that are definitely better than these two, and if could point me a link to them, would be great but, exactly between those two, which one would you consider to be better, regarding both performance and quality/precision? Thanks in advance.

    Read the article

  • What constitutes proper use of threads in programming?

    - by Smith
    I am tired of hearing people recommend that you should use only one thread per process, while many programs use up to 100 per process! take for example some common programs vb.net ide uses about 25 thread when not debugging System uses about 100 chrome uses about 19 Avira uses more than about 50 Any time I post a thread related question, I am reminded almost every time that I should not use more that one thread per process, and all the programs I mention above are ruining on my system with a single processor. What constitutes proper use of threads in programming? Please make general comment, but I'd prefer .NET framework thanks EDIT changed processor to process

    Read the article

  • Cheaper alternatives to 99Designs.com (outsource CSS design)

    - by Chris Smith
    I'm designing my own website as a side project and I want the site to look professional. (Read, not designed by a programmer.) I don't mind spending a little money to have a professional do it, but design sites like 99designs.com cost way to much. (~$500+) Is there a cheaper (~$100 - $200) alternative for getting a designer to improve an existing site? (Things like updating the CCS or suggesting better ways for laying out the navigation.) Or is my best bet trying to pick up a freelancer on Craigslist?

    Read the article

  • Weekly Cloud Roundup 2012-15

    - by Alan Smith
    Filtering the informative, insightful and quirky from the fire hose of cloud-based hype. Irving Wladawsky-Berger provides some great insight into The Complex Transition to the Cloud, sharing his views on the slow adoption of cloud computing in organizations. “…a prediction by the research firm Gartner that while cloud computing will continue to grow at almost 20 percent a year, it will account for less than 5 percent of totally IT spending in 2015.” With a more positive mindset, Balaji Viswanathan highlights 7 Salient Trends and Directions in Cloud Computing that could be shaping the industry over the next few years. Cloud computing also looks to save energy “A small business with 100 users that moved the Microsoft applications to the cloud could cut energy use and carbon emissions by 90%. Large organizations with 10,000 users saw a 30% reduction.” More on that story here. The expansion of Windows Azure has been in the news with the announcement of “East US” and “West US” datacenters; this was covered by Visual Studio Magazine and Mary-Jo, and according to thenextweb.com Microsoft are also building $112 million data center in Wyoming. The cloud price war is still in full swing with Joe Panettieri discussing the pricing of Windows Azure and Office 365 and asking How Low Can It Go?

    Read the article

  • Should I expect my team to have more than a basic proficiency with our source control system?

    - by Joshua Smith
    My company switched from Subversion to Git about three months ago. We had weeks of advance notice prior to the switch. Since I'd never used Git before (or any other DVCS), I read Pro Git and spent a little time spinning up my own repositories and playing around, so that when we switched I'd be able to keep working with minimal pain. Now I'm the 'Git guy' by default. With a couple of exceptions, most of my team still has no idea how Git works. For example, they still think of branches as complete copies of the source code, and even go so far as to clone the repo into multiple folders (one per branch). They generally look at Git as a scary black box. Given the fundamental nature of source control in our daily work (not to mention the ridiculous amount of power Git affords us), I'm of the opinion that any dev who doesn't achieve a certain level of proficiency with it is a liability. Should I expect my team to have at least some understanding of how Git works internally, and how to use it beyond the most basic pull/merge/push operations? Or am I just making something out of nothing?

    Read the article

  • VS2012 Light or Dark Theme Programmatically

    - by Neil Smith
    HKEY_CURRENT_USER\Software\Microsoft\VisualStudio\11.0\General\CurrentTheme\ Contains a guid (in the case of dark 1ded0138-47ce-435e-84ef-9ec1f439b749 This guid maps to a list of keys under HKEY_CURRENT_USER\Software\Microsoft\VisualStudio\11.0_Config\Themes\ 1ded0138-47ce-435e-84ef-9ec1f439b749 = Dark a5c004b4-2d4b-494e-bf01-45fc492522c7 = High Contrast de3dbbcd-f642-433c-8353-8f1df4370aba = Light

    Read the article

  • What happened this type of naming convention?

    - by Smith
    I have read so many docs about naming conventions, most recommending both Pascal and Camel naming conventions. Well, I agree to this, its ok. This might not be pleasing to some, but I am just trying to get you opinion why you name you objects and classes in a certain way. What happened to this type of naming conventions, or why are they bad? I want to name a struct, and i prefix it with struct. My reason, so that in IntelliSense, I see all the struct in one place, and anywhere I see the struct prefix, I know it's a struct: structPerson structPosition anothe example is the enum, although I may not prefix it with "enum", but maybe with "enm": enmFruits enmSex again my reason is so that in IntelliSense, I see all my enums in one place. Because, .NET has so many built in data structures, I think this helps me do less searching. Please I used .NET in this example, but I welcome language agnostic answers.

    Read the article

  • Cheaper alternatives to 99Designs.com (outsource CSS design)

    - by Chris Smith
    I'm designing my own website as a side project and I want the site to look professional. (Read, not designed by a programmer.) I don't mind spending a little money to have a professional do it, but design sites like 99designs.com cost way to much. (~$500+) Is there a cheaper (~$100 - $200) alternative for getting a designer to improve an existing site? (Things like updating the CCS or suggesting better ways for laying out the navigation.) Or is my best bet trying to pick up a freelancer on Craigslist?

    Read the article

  • When is the default storage rule not really the default storage rule?

    - by Kevin Smith
    In 11g WebCenter Content (WCC) introduced dispersion rules in the vault and weblayout directory paths to better distribute content across the directories. The dispersion rule was based on dRevClassID. The only problem with this is that dRevClassID did not remain the same when you copied content from one WCC instance to another using Archiver like in a contribution-consumption scenario. This could cause problems because the web-viewable path would not be the same between the contribution and consumption instances. In the PS5 (11.1.1.6.0) release of WCC they addressed this by configuring the File Store Provider (FSP) so that all new content would use a storage rule with a dispersion rule based on dDocName, which would stay the same when content was copied to another WCC instance. To support migration from older versions of WCC they left the default storage rule unchanged and created a new storage rule called DispByContentId and made that the default storage rule for all new content. I only stumbled upon this a while back when I was trying to change the FSP configuration so that all content used a webless storage rule. I changed the default storage rule, restarted WCC, and checked in a new content item. To my surprise the new content was not created as webless. I struggled with this for a while until I noticed there were multiple storage rules defined in the FSP configuration. When I looked at the default value for the xStorageRule field in Configuration Manager, sure enough it was no longer default, but was now DispByContentId. Once I updated the DispByContentId storage rule to webless and restarted WCC all my new content was now created using the webless storage rule, just like I wanted. I noticed when I was creating this blog post that the default storage rule is also listed on the File Store Provider Information page, but I guess I didn't see that when I originally did this.

    Read the article

< Previous Page | 108 109 110 111 112 113 114 115 116 117 118 119  | Next Page >