Search Results

Search found 93838 results on 3754 pages for 'aspire one'.

Page 411/3754 | < Previous Page | 407 408 409 410 411 412 413 414 415 416 417 418  | Next Page >

  • Cluster Nodes as RAID Drives

    - by BuckWoody
    I'm unable to sleep tonight so I thought I would push this post out VERY early. When you don't sleep your mind takes interesting turns, which can be a good thing. I was watching a briefing today by a couple of friends as they were talking about various ways to arrange a Windows Server Cluster for SQL Server. I often see an "active" node of a cluster with a "passive" node backing it up. That means one node is working and accepting transactions, and the other is not doing any work but simply "standing by" waiting for the first to fail over. The configuration in the demonstration I saw was a bit different. In this example, there were three nodes that were actively working, and a fourth standing by for all three. I've put configurations like this one into place before, but as I was looking at their architecture diagram, it looked familar - it looked like a RAID drive setup! And that's not a bad way to think about your cluster arrangements. The same concerns you might think about for a particular RAID configuration provides a good way to think about protecting your systems in general. So even if you're not staying awake all night thinking about SQL Server clusters, take this post as an opportunity for "lateral thinking" - a way of combining in your mind the concepts from one piece of knowledge to another. You might find a new way of making your technical environment a little better. Share this post: email it! | bookmark it! | digg it! | reddit! | kick it! | live it!

    Read the article

  • Procedural... house with rooms generator

    - by pek
    I've been looking at some algorithms and articles about procedurally generating a dungeon. The problem is, I'm trying to generate a house with rooms, and they don't seem to fit my requirements. For one, dungeons have corridors, where houses have halls. And while initially they might seem the same, a hall is nothing more than the area that isn't a room, whereas a corridor is specifically designed to connect one area to another. Another important difference with a house is that you have a specific width and height, and you have to fill the entire thing with rooms and halls, whereas with a dungeon, there is empty space. I think halls in a house is something in between a dungeon corridor (gets you to other rooms) and an empty space in the dungeon (it's not explicitly defined in code). More specifically, the requirements are: There is a set of predefined rooms I cannot create walls and doors on the fly. Rooms can be rotated but not resized Again, because I have a predefined set of rooms, I can only rotate them, not resize them. The house dimensions are set and has to be entirely filled with rooms (or halls) I.e. I want to fill a 14x20 house with the available rooms making sure there is no empty space. Here are some images to make this a little more clear: As you can see, in the house, the "empty space" is still walkable and it gets you from one room to another. So, having said all this, maybe a house is just a really really tightly packed dungeon with corridors. Or it's something easier than a dungeon. Maybe there is something out there and I haven't found it because I don't really know what to search for. This is where I'd like your help: could you give me pointers on how to design this algorithm? Any thoughts on what steps it will take? If you have created a dungeon generator, how would you modify it to fit my requirements? You can be as specific or as generic as you like. I'm looking to pick your brains, really.

    Read the article

  • PowerShell to fetch a SQL Execution Plan

    - by Rob Farley
    With PowerShell becoming the scripting language of choice for many people, I’ve occasionally wondered about using it to analyse execution plans. After all, an execution plan is just XML, and PowerShell is just one tool which will very easily handle xml. The thing is – there’s no Get-SqlPlan cmdlet available, which has frustrated me in the past. Today I figured I’d make one. I know that I can write T-SQL to get an execution plan using SET SHOWPLAN_XML ON, but the problem is that this must be the only statement in a batch. So I used go, and a couple of newlines, and whipped up the following one-liner: function Get-SqlPlan([string] $query, [string] $server, [string] $db) { return ([xml] (invoke-sqlcmd -Server $server -Database $db -Query "set showplan_xml on;`ngo`n$query").Item( 0)) } (but please bear in mind that I have the SQL Snapins installed, which provides invoke-sqlcmd) To use this, I just do something like: $plan = get-sqlplan "select name from Production.Product" "." "AdventureWorks" And then find myself with an easy way to navigate through an execution plan! At some point I should make the function more robust, but this should be a good starter for any SQL PowerShell enthusiasts (like Aaron Nelson) out there.

    Read the article

  • Difference between Detach/Attach and Restore/BackUp a DB

    - by SAMIR BHOGAYTA
    Transact-SQL BACKUP/RESTORE is the normal method for database backup and recovery. Databases can be backed up while online. The backup file size is usually smaller than the database files since only used pages are backed up. Also, in the FULL or BULK_LOGGED recovery model, you can reduce potential data loss by performing transaction log backups. Detaching a database removes the database from SQL Server while leaving the physical database files intact. This allows you to rename or move the physical files and then re-attach. Although one could perform cold backups using this technique, detach/attach isn't really intended to be used as a backup/recovery process. Commonly it is recommended that you use BACKUP/RESTORE for disaster recovery (DR) scenario and copying data from one location to another. But this is not absolute, sometimes for a very large database, if you want to move it from one location to another, backup/restore process may spend a lot of time which you do not like, in this case, detaching/attaching a database is a better way since you can attach a workable database very fast. But you need to aware that detaching a database will bring it offline for a short time and detaching/attaching does not provide DR function. For more information about detaching and attaching databases, you can refer to: Detaching and Attaching Databases http://technet.microsoft.com/en-us/library/ms190794.aspx

    Read the article

  • Using a portable USB monitor in Ubuntu 13.04 (AOC e1649Fwu - DisplayLink)

    Having access to a little bit of IT hardware extravaganza isn't that easy here in Mauritius for exactly two reasons - either it is simply not available or it is expensive like nowhere. Well, by chance I came across an advert by a local hardware supplier and their offer of the week caught my attention - a portable USB monitor. Sounds cool, and the specs are okay as well. It's completely driven via USB 2.0, has a light weight, the dimensions would fit into my laptop bag and the resolution of 1366 x 768 pixels is okay for a second screen. Long story, short ending: I called them and only got to understand that they are out of stock - how convenient! Well, as usual I left some contact details and got the regular 'We call you back' answer. Surprisingly, I didn't receive a phone call as promised and after starting to complain via social media networks they finally came back to me with new units available - and *drum-roll* still the same price tag as promoted (and free delivery on top as one of their employees lives in Flic en Flac). Guess, it was a no-brainer to get at least one unit to fool around with. In worst case it might end up as image frame on the shelf or so... The usual suspects... Ubuntu first! Of course, the packing mentions only Windows or Mac OS as supported operating systems and without hesitation at all, I hooked up the device on my main machine running on Ubuntu 13.04. Result: Blackout... Hm, actually not the situation I was looking for but okay can't be too difficult to get this piece of hardware up and running. Following the output of syslogd (or dmesg if you prefer) the device has been recognised successfully but we got stuck in the initialisation phase. Oct 12 08:17:23 iospc2 kernel: [69818.689137] usb 2-4: new high-speed USB device number 5 using ehci-pciOct 12 08:17:23 iospc2 kernel: [69818.800306] usb 2-4: device descriptor read/64, error -32Oct 12 08:17:24 iospc2 kernel: [69819.043620] usb 2-4: New USB device found, idVendor=17e9, idProduct=4107Oct 12 08:17:24 iospc2 kernel: [69819.043630] usb 2-4: New USB device strings: Mfr=1, Product=2, SerialNumber=3Oct 12 08:17:24 iospc2 kernel: [69819.043636] usb 2-4: Product: e1649FwuOct 12 08:17:24 iospc2 kernel: [69819.043642] usb 2-4: Manufacturer: DisplayLinkOct 12 08:17:24 iospc2 kernel: [69819.043647] usb 2-4: SerialNumber: FJBD7HA000778Oct 12 08:17:24 iospc2 kernel: [69819.046073] hid-generic 0003:17E9:4107.0008: hiddev0,hidraw5: USB HID v1.10 Device [DisplayLink e1649Fwu] on usb-0000:00:1d.7-4/input1Oct 12 08:17:24 iospc2 mtp-probe: checking bus 2, device 5: "/sys/devices/pci0000:00/0000:00:1d.7/usb2/2-4"Oct 12 08:17:24 iospc2 mtp-probe: bus: 2, device: 5 was not an MTP deviceOct 12 08:17:30 iospc2 kernel: [69825.411220] [drm] vendor descriptor length:17 data:17 5f 01 00 15 05 00 01 03 00 04Oct 12 08:17:30 iospc2 kernel: [69825.498778] udl 2-4:1.0: fb1: udldrmfb frame buffer deviceOct 12 08:17:30 iospc2 kernel: [69825.498786] [drm] Initialized udl 0.0.1 20120220 on minor 1Oct 12 08:17:30 iospc2 kernel: [69825.498909] usbcore: registered new interface driver udl The device has been recognised as USB device without any question and it is listed properly: # lsusb...Bus 002 Device 005: ID 17e9:4107 DisplayLink ... A quick and dirty research on the net gave me some hints towards the udlfb framebuffer device for USB DisplayLink devices. By default this kernel module is blacklisted $ less /etc/modprobe.d/blacklist-framebuffer.conf | grep udl#blacklist udlblacklist udlfb and it is recommended to load it manually. So, unloading the whole udl stack and giving udlfb a shot: Oct 12 08:22:31 iospc2 kernel: [70126.642809] usbcore: registered new interface driver udlfb But still no reaction on the external display which supposedly should have been on and green. Display okay? Test run on Windows Just to be on the safe side and to exclude any hardware related defects or whatsoever - you never know what happened during delivery. I moved the display to a new position on the opposite side of my laptop, installed the display drivers first in Windows Vista (I know, I know...) as recommended in the manual, and then finally hooked it up on that machine. Tada! Display has been recognised correctly and I have a proper choice between cloning and extending my desktop. Testing whether the display is working properly - using Windows Vista Okay, good to know that there is nothing wrong on the hardware side just software... Back to Ubuntu - Kernel too old Some more research on Google and various hits recommend that the original displaylink driver has been merged into the recent kernel development and one should manually upgrade the kernel image (and both header) packages for Ubuntu. At least kernel 3.9 or higher would be necessary, and so I went out to this URL: http://kernel.ubuntu.com/~kernel-ppa/mainline/ and I downloaded all the good stuff from the v3.9-raring directory. The installation itself is easy going via dpkg: $ sudo dpkg -i linux-image-3.9.0-030900-generic_3.9.0-030900.201304291257_amd64.deb$ sudo dpkg -i linux-headers-3.9.0-030900_3.9.0-030900.201304291257_all.deb$ sudo dpkg -i linux-headers-3.9.0-030900-generic_3.9.0-030900.201304291257_amd64.deb As with any kernel upgrades it is necessary to restart the system in order to use the new one. Said and done: $ uname -r3.9.0-030900-generic And now connecting the external display gives me the following output in /var/log/syslog: Oct 12 17:51:36 iospc2 kernel: [ 2314.984293] usb 2-4: new high-speed USB device number 6 using ehci-pciOct 12 17:51:36 iospc2 kernel: [ 2315.096257] usb 2-4: device descriptor read/64, error -32Oct 12 17:51:36 iospc2 kernel: [ 2315.337105] usb 2-4: New USB device found, idVendor=17e9, idProduct=4107Oct 12 17:51:36 iospc2 kernel: [ 2315.337115] usb 2-4: New USB device strings: Mfr=1, Product=2, SerialNumber=3Oct 12 17:51:36 iospc2 kernel: [ 2315.337122] usb 2-4: Product: e1649FwuOct 12 17:51:36 iospc2 kernel: [ 2315.337127] usb 2-4: Manufacturer: DisplayLinkOct 12 17:51:36 iospc2 kernel: [ 2315.337132] usb 2-4: SerialNumber: FJBD7HA000778Oct 12 17:51:36 iospc2 kernel: [ 2315.338292] udlfb: DisplayLink e1649Fwu - serial #FJBD7HA000778Oct 12 17:51:36 iospc2 kernel: [ 2315.338299] udlfb: vid_17e9&pid_4107&rev_0129 driver's dlfb_data struct at ffff880117e59000Oct 12 17:51:36 iospc2 kernel: [ 2315.338303] udlfb: console enable=1Oct 12 17:51:36 iospc2 kernel: [ 2315.338306] udlfb: fb_defio enable=1Oct 12 17:51:36 iospc2 kernel: [ 2315.338309] udlfb: shadow enable=1Oct 12 17:51:36 iospc2 kernel: [ 2315.338468] udlfb: vendor descriptor length:17 data:17 5f 01 0015 05 00 01 03 00 04Oct 12 17:51:36 iospc2 kernel: [ 2315.338473] udlfb: DL chip limited to 1500000 pixel modesOct 12 17:51:36 iospc2 kernel: [ 2315.338565] udlfb: allocated 4 65024 byte urbsOct 12 17:51:36 iospc2 kernel: [ 2315.343592] hid-generic 0003:17E9:4107.0009: hiddev0,hidraw5: USB HID v1.10 Device [DisplayLink e1649Fwu] on usb-0000:00:1d.7-4/input1Oct 12 17:51:36 iospc2 mtp-probe: checking bus 2, device 6: "/sys/devices/pci0000:00/0000:00:1d.7/usb2/2-4"Oct 12 17:51:36 iospc2 mtp-probe: bus: 2, device: 6 was not an MTP deviceOct 12 17:51:36 iospc2 kernel: [ 2315.426583] udlfb: 1366x768 @ 59 Hz valid modeOct 12 17:51:36 iospc2 kernel: [ 2315.426589] udlfb: Reallocating framebuffer. Addresses will change!Oct 12 17:51:36 iospc2 kernel: [ 2315.428338] udlfb: 1366x768 @ 59 Hz valid modeOct 12 17:51:36 iospc2 kernel: [ 2315.428343] udlfb: set_par mode 1366x768Oct 12 17:51:36 iospc2 kernel: [ 2315.430620] udlfb: DisplayLink USB device /dev/fb1 attached. 1366x768 resolution. Using 4104K framebuffer memory Okay, that's looks more promising but still only blackout on the external screen... And yes, due to my previous modifications I swapped the blacklisted kernel modules: $ less /etc/modprobe.d/blacklist-framebuffer.conf | grep udlblacklist udl#blacklist udlfb Silly me! Okay, back to the original situation in which udl is allowed and udlfb blacklisted. Now, the logging looks similar to this and the screen shows those maroon-brown and azure-blue horizontal bars as described on other online resources. Oct 15 21:27:23 iospc2 kernel: [80934.308238] usb 2-4: new high-speed USB device number 5 using ehci-pciOct 15 21:27:23 iospc2 kernel: [80934.420244] usb 2-4: device descriptor read/64, error -32Oct 15 21:27:24 iospc2 kernel: [80934.660822] usb 2-4: New USB device found, idVendor=17e9, idProduct=4107Oct 15 21:27:24 iospc2 kernel: [80934.660832] usb 2-4: New USB device strings: Mfr=1, Product=2, SerialNumber=3Oct 15 21:27:24 iospc2 kernel: [80934.660838] usb 2-4: Product: e1649FwuOct 15 21:27:24 iospc2 kernel: [80934.660844] usb 2-4: Manufacturer: DisplayLinkOct 15 21:27:24 iospc2 kernel: [80934.660850] usb 2-4: SerialNumber: FJBD7HA000778Oct 15 21:27:24 iospc2 kernel: [80934.663391] hid-generic 0003:17E9:4107.0008: hiddev0,hidraw5: USB HID v1.10 Device [DisplayLink e1649Fwu] on usb-0000:00:1d.7-4/input1Oct 15 21:27:24 iospc2 mtp-probe: checking bus 2, device 5: "/sys/devices/pci0000:00/0000:00:1d.7/usb2/2-4"Oct 15 21:27:24 iospc2 mtp-probe: bus: 2, device: 5 was not an MTP deviceOct 15 21:27:25 iospc2 kernel: [80935.742407] [drm] vendor descriptor length:17 data:17 5f 01 00 15 05 00 01 03 00 04Oct 15 21:27:25 iospc2 kernel: [80935.834403] udl 2-4:1.0: fb1: udldrmfb frame buffer deviceOct 15 21:27:25 iospc2 kernel: [80935.834416] [drm] Initialized udl 0.0.1 20120220 on minor 1Oct 15 21:27:25 iospc2 kernel: [80935.836389] usbcore: registered new interface driver udlOct 15 21:27:25 iospc2 kernel: [80936.021458] [drm] write mode info 153 Next, it's time to enable the display for our needs... This can be done either via UI or console, just as you'd prefer it. Adding the external USB display under Linux isn't an issue after all... Settings Manager => Display Personally, I like the console. With the help of xrandr we get the screen identifier first $ xrandrScreen 0: minimum 320 x 200, current 3200 x 1080, maximum 32767 x 32767LVDS1 connected 1280x800+0+0 (normal left inverted right x axis y axis) 331mm x 207mm...DVI-0 connected 1366x768+0+0 (normal left inverted right x axis y axis) 344mm x 193mm   1366x768       60.0*+ and then give it the usual shot with auto-configuration. Let the system decide what's best for your hardware... $ xrandr --output DVI-0 --off$ xrandr --output DVI-0 --auto And there we go... Cloned output of main display: New kernel, new display... The external USB display works out-of-the-box with a Linux kernel > 3.9.0. Despite of a good number of resources it is absolutely not necessary to create a Device or Screen section in one of Xorg.conf files. This information belongs to the past and is not valid on kernel 3.9 or higher. Same hardware but Windows 8 Of course, I wanted to know how the latest incarnation from Redmond would handle the new hardware... Flawless! Most interesting aspect here: I did not use the driver installation medium on purpose. And I was right... not too long afterwards a dialog with the EULA of DisplayLink appeared on the main screen. And after confirmation of same it took some more seconds and the external USB monitor was ready to rumble. Well, and not only that one... but see for yourself. This time Windows 8 was the easiest solution after all. Resume I can highly recommend this type of hardware to anyone asking me. Although, it's dimensions are 15.6" it is actually lighter than my Samsung Galaxy Tab 10.1 and it still fits into my laptop bag without any issues. From now on... no more single screen while developing software on the road!

    Read the article

  • Python web frameworks comparisons

    - by stupidLearner
    I recently asked a question on SO about Python web frameworks: http://stackoverflow.com/questions/4909306/python-web-frameworks-vs-java-web-frameworks-how-is-web-development-in-python-do I want to learn one just for fun but it also has to be able to help me deliver a proper working application. I am looking for a framework with lots of features, powerful, mature, with large community, good documentation, books etc. I need something that will help me be more productive in developing my app and not waste time figuring out how to do a certain thing in the framework or how to write workaround around the limitations of the framework. I was thinking one of the following: django, zope, turbogears, pylons. Off course the war is raging out there and there are other alternatives but seems Django is at the top... or is it just hype? I am interested in pros and cons of each. What was the best feature you think the framework has? What is the thing it lacks? What could have been done differently. Help me chose one to learn for starters.

    Read the article

  • What are benefit/drawbacks of classifying defects during a peer code review

    - by DXM
    About 3 months ago, our engineering group rolled out Review Board to be used for all peer code reviews. Today, I had a discussion with one of the people involved in that process and found out that we are already looking for a replacement (possibly something commercial) because of several missing features. One of the features that is apparently asked by many people is the ability to classify/categorize each code review comment (i.e. is it a style issue, coding convention, resource leak, logic error, crash... whatever). For those teams that regularly practice code review, is this categorization a common practice? Do you do it? have you done it in the past? Is it good/bad? On one hand, it gives the team some more metrics and possibly will indicate more specific areas where developers may potentially need to be trained in (at least that seems to be the argument). Are there other benefits? And on the other hand, and this is my concern, is that it will slow down code review process that much more. As a team lead, I've done a fairly large share of reviews, and I've always liked the ability, to highlight a chunk of code, hammer off a comment and move on as fast as possible. Although I haven't tried it personally, I have a feeling that expanding that combo box every time and scrolling/searching for the right category would feel like something is tripping you. Also if we start keeping metrics on this stuff, my other concern is that valuable code review meeting time will be spent on arguing whether something is a logic error or if it should be classified as a crash.

    Read the article

  • MVC 4 and the App_Start folder

    - by pjohnson
    I've been delving into ASP.NET MVC 4 a little since its release last month. One thing I was chomping at the bit to explore was its bundling and minification functionality, for which I'd previously used Cassette, and been fairly happy with it. MVC 4's functionality seems very similar to Cassette's; the latter's CassetteConfiguration class matches the former's BundleConfig class, specified in a new directory called App_Start.At first glance, this seems like another special ASP.NET folder, like App_Data, App_GlobalResources, App_LocalResources, and App_Browsers. But Visual Studio 2010's lack of knowledge about it (no Solution Explorer option to add the folder, nor a fancy icon for it) made me suspicious. I found the MVC 4 project template has five classes there--AuthConfig, BundleConfig, FilterConfig, RouteConfig, and WebApiConfig. Each of these is called explicitly in Global.asax's Application_Start method. Why create separate classes, each with a single static method? Maybe they anticipate a lot more code being added there for large applications, but for small ones, it seems like overkill. (And they seem hastily implemented--some declared as static and some not, in the base namespace instead of an App_Start/AppStart one.) Even for a large application I work on with a substantial amount of code in Global.asax.cs, a RouteConfig might be warranted, but the other classes would remain tiny.More importantly, it appears App_Start has no special magic like the other folders--it's just convention. I found it first described in the MVC 3 timeframe by Microsoft architect David Ebbo, for the benefit of NuGet and WebActivator; apparently some packages will add their own classes to that directory as well. One of the first appears to be Ninject, as most mentions of that folder mention it, and there's not much information elsewhere about this new folder.

    Read the article

  • Why are SW engineering interviews disproportionately difficult?

    - by stackoverflowuser2010
    First, some background on me. I have a PhD in CS and have had jobs both as a software engineer and as an R&D research scientist, both at Very Large Corporations You Know Very Well. I recently changed jobs and interviewed for both types of jobs (as I have done in the past). My observation: SW engineer job interviews are way, way disproportionately more difficult than CS researcher job interviews, but the researcher job is higher paying, more competitive, more rewarding, more interesting, and has a higher upside. Here's a typical interview loop for researcher: Phone interview to see if my research is in alignment with the lab's researcher In-person, give presentation on my recent research for one hour (which represents maybe 9 month's worth of work), answer questions In-person one-on-one interviews with about 5 researchers, where they ask me very reasonable questions on my work/publications/patents, including: technical questions, where my work fits into related work, and how I can extend my work to new areas Here's a typical interview loop for SW engineer: Phone interview where I'm asked algorithm questions and maybe do some coding. Pretty standard. In-person interviews at the whiteboard where they drill the F*** out of you on esoteric C++ minutia (e.g. how does a polymorphic virtual function call work), algorithms (make all-pairs-shortest-path algorithm work for 1B vertices), system design (design a database load balancer), etc. This goes on for six or seven interviews. Ridiculous. Why would anyone be willing to put up with this? What is the point of asking about C++ trivia or writing code to prove yourself? Why not make the SE interview more like the researcher interview where you give a talk about what you've done? How are technical job interviews for other fields, like physics, chemistry, civil engineering, mechanical engineering?

    Read the article

  • High-Level Application Architecture Question

    - by Jesse Bunch
    So I'm really wanting to improve how I architect the software I code. I want to focus on maintainability and clean code. As you might guess, I've been reading a lot of resources on this topic and all it's doing is making it harder for me to settle on an architecture because I can never tell if my design is the one that the more experienced programmer would've chosen. So I have these requirements: I should connect to one vendor and download form submissions from their API. We'll call them the CompanyA. I should then map those submissions to a schema fit for submitting to another vendor for integration with the email service provider. We'll call them the CompanyB. I should then submit those responses to the ESP (CompanyB) and then instruct the ESP to send that submitter an email. So basically, I'm copying data from one web service to another and then performing an action at the latter web service. I've identified a couple high-level services: The service that downloads data from CompanyA. I called this the CompanyAIntegrator. The service that submits the data to CompanyB. I called this CompanyBIntegrator. So my questions are these: Is this a good design? I've tried to separate the concerns and am planning to use the facade pattern to make the integrators interchangeable if the vendors change in the future. Are my naming conventions accurate and meaningful to you (who knows nothing specific of the project)? Now that I have these services, where should I do the work of taking output from the CompanyAIntegrator and getting it in the format for input to the CompanyBIntegrator? Is this OK to be done in main()? Do you have any general pointers on how you'd code something like this? I imagine this scenario is common to us engineers---especially those working in agencies. Thanks for any help you can give. Learning how to architect well is really mind-cluttering.

    Read the article

  • Interim Update #1: Microsoft Office 2010 and E-Business Suite

    - by Steven Chan
    Congratulations to my colleagues at Microsoft on their launch of Microsoft Office 2010 yesterday.  Questions about our certification plans for Office 2010 are filling my inbox, so here's an interim update on our plans.  If you've reached this article via a search engine, it's possible that a later update on our status is available.  For our latest status, please check the Desktop Client Certifications section of our one-page Certifications summary.Our current plans for Office 2010We plan to certify Oracle E-Business Suite Release 11i and 12 with Microsoft Office 2010.When will Office 2010 be certified with EBS?Oracle's Revenue Recognition rules prohibit us from discussing certification and release dates, but you're welcome to monitor or subscribe to this blog for updates, which I'll post as soon as soon as they're available.    How does the E-Business Suite work with Microsoft Office?The Oracle E-Business Suite is comprised of several product families such as Financials, Supply Chain Management, Human Resources, and so on.  These product families group together collections of individual products.  Some of these products may be optionally integrated with one or more Microsoft Office components such as Excel, Word, and Projects.Individual E-Business Suite product teams have the option of building integrations between their products and one or more Microsoft Office components.  This is not mandatory.  Over forty E-Business Suite teams offer these kinds of Office integrations today.

    Read the article

  • Codeigniter + JQuery + Processing.js to replace a Delphi App

    - by Peter Turner
    So, I've got a mandate to make our aged trillion lined Delphi app web based and it needs to make heavy use of the <canvas> element (HTML5 compatibility doesn't seem to be a big issue since we can just make our clients use a compatible browser the way we'd make them use a compatible version of Windows in the win32 environment). The Delphi app in question is almost completely database driven and will still pretty much continue to be developed as the main product. What I am tasked with is pretty much recreating a scaled down version of the program that performs the major functions of the whole program. I couldn't find any frameworks that simulate windows forms using the canvas element, I'm assuming this is probably by design since it is easier just to use HTML, well, be that as it may, I still think it would be cool to have a few of my cool controls on the web (TRichView and TVirtualTree, etc...) So my question is, to anyone who has tried this before, A.) What can we use for an IDE to code this web app (I just use emacs, but no one else in my company does)? B.) Is it a good idea to mix PHP and Processing.JS? It seems like I'm using a lot of AJAX to get anything to happen. 3 calls just for one dialog box to pop up, Loads the HTML for the dialog, Loads the XML to populate the database info on the form Loads the processing.js PJS file which draws the database info to the canvas. Is three a lot, do people usually combine all their gets into one?

    Read the article

  • How can I tell if a replacement display is compatible with my laptop?

    - by moraleida
    I have an Acer Aspire 4810TZ from 2010 which I truly love - it's a 3 year old notebook that still carries on for about 5 to 6h straight off the battery. I've also invested in it with an SSD and extra RAM and it's fine for my use right now. I'd really rather not just dump it. Now... its screen is dead and I'm looking for replacement parts, but it turns out to be a kind of expensive display - AUO B140XW02 14" LED. US$ 125 was the cheapest I found here (Rio de Janeiro, Brazil). The official dealer sells them for US$ 175. I'm aware of How do I know if a LCD is compatible with my Laptop? and Can I replace a laptop screen with one of a different resolution? but the answers don't really help me in this case. Also Replacing an LCD screen in a laptop gives me a hint but I'm completely dumb at hardware so this is why i'm asking: Is it possible to buy an LCD or a lower grade LED screen with the same connectors? What characteristics should I look for to find out if a certain display is compatible?

    Read the article

  • Since a few days, Dropbox stopped syncing

    - by Fabio
    Theres a new problem with nautilus-dropbox and is the worst for a service like Dropbox: stopped syncing and their tech support don't believe their users. The problem is: start dropbox it downloads index creates folders start to download files stop downloading files or the speed is like 0.3k/seg even deleting everything doesn't work, even syncing a single folder with a few files, not even configuring a different limit speed or let all your bandwith, nothing works for me and i have found many users with the same problem. it's strange, i have two computers with ubuntu 14.04 and one with windows 7, and in one of the linux machines it works fine, but not in the other one. Both of them are x64. The answers from the tech support starts with "delete everything and reinstall" to "you may be blocking port 443", yeah, sure (and if i do that half of internet dissapears, idiot), but nothing works. Someone has some idea? i cant find any log file (at least a text log file like in /var/log) to understand what is dropbox doing and what is not working. PS: sorry for my "english", is not my language Examples: dropbox slow on ubuntu 14.04 Dropbox does not sync on ubuntu 14.04 Dropbox Status (in spanish but is the same thing) fabio@canopus:~$ dropbox status Sincronizando (Quedan 330 archivos., Faltan 9 días.) Descargando 330 archivos... (0,3 KB/s, Faltan 9 días.) 9 days for 330 files / 100Mb

    Read the article

  • Organizing Connections with Folders in Oracle SQL Developer

    - by thatjeffsmith
    How many Oracle databases do you work with on a regular basis? I’m guessing the answer for most of you lies between 1 and 500. This post is really geared for those of you who deal with more than just a handful (5) of database connections. Filters are nice when you need to work with a subset of table data, or even a list of tables. So why wouldn’t they be just as useful for organizing your connections? Here’s my complete list of databases: The folders aren’t there by default, you add them as you need them. Now this isn’t an overly large connection list. But when I need to fire up an impromptu demo for a customer, it’s very nice to be able to drill down into JUST those ‘safe’ environments. This actually saves me a few seconds every time I need to connect to one of my databases. So while it’s a very simple feature, it’s one of those things that I recommend EVERYONE take advantage of as it will save them hours of time over the long haul. Easier to find means I get to work a few seconds faster. This also helps me from making mistakes in ‘production’ environments! How to Add a Connection Folder Select a connection you want to organize. Mouse-right-click, and choose ‘Add to folder.’ You can throw it into a new container or an existing one. Lather, rinse, and repeat as necessary. The only trick is remembering to right-click! Special thanks to @dresendi for today’s topic! He asked how to do this and I realized I hadn’t blogged the topic yet

    Read the article

  • Code Reuse is (Damn) Hard

    - by James Michael Hare
    Being a development team lead, the task of interviewing new candidates was part of my job.  Like any typical interview, we started with some easy questions to get them warmed up and help calm their nerves before hitting the hard stuff. One of those easier questions was almost always: “Name some benefits of object-oriented development.”  Nearly every time, the candidate would chime in with a plethora of canned answers which typically included: “it helps ease code reuse.”  Of course, this is a gross oversimplification.  Tools only ease reuse, its developers that ultimately can cause code to be reusable or not, regardless of the language or methodology. But it did get me thinking…  we always used to say that as part of our mantra as to why Object-Oriented Programming was so great.  With polymorphism, inheritance, encapsulation, etc. we in essence set up the concepts to help facilitate reuse as much as possible.  And yes, as a developer now of many years, I unquestionably held that belief for ages before it really struck me how my views on reuse have jaded over the years.  In fact, in many ways Agile rightly eschews reuse as taking a backseat to developing what's needed for the here and now.  It used to be I was in complete opposition to that view, but more and more I've come to see the logic in it.  Too many times I've seen developers (myself included) get lost in design paralysis trying to come up with the perfect abstraction that would stand all time.  Nearly without fail, all of these pieces of code become obsolete in a matter of months or years. It’s not that I don’t like reuse – it’s just that reuse is hard.  In fact, reuse is DAMN hard.  Many times it is just a distraction that eats up architect and developer time, and worse yet can be counter-productive and force wrong decisions.  Now don’t get me wrong, I love the idea of reusable code when it makes sense.  These are in the few cases where you are designing something that is inherently reusable.  The problem is, most business-class code is inherently unfit for reuse! Furthermore, the code that is reusable will often fail to be reused if you don’t have the proper framework in place for effective reuse that includes standardized versioning, building, releasing, and documenting the components.  That should always be standard across the board when promoting reusable code.  All of this is hard, and it should only be done when you have code that is truly reusable or you will be exerting a large amount of development effort for very little bang for your buck. But my goal here is not to get into how to reuse (that is a topic unto itself) but what should be reused.  First, let’s look at an extension method.  There’s many times where I want to kick off a thread to handle a task, then when I want to reign that thread in of course I want to do a Join on it.  But what if I only want to wait a limited amount of time and then Abort?  Well, I could of course write that logic out by hand each time, but it seemed like a great extension method: 1: public static class ThreadExtensions 2: { 3: public static bool JoinOrAbort(this Thread thread, TimeSpan timeToWait) 4: { 5: bool isJoined = false; 6:  7: if (thread != null) 8: { 9: isJoined = thread.Join(timeToWait); 10:  11: if (!isJoined) 12: { 13: thread.Abort(); 14: } 15: } 16: return isJoined; 17: } 18: } 19:  When I look at this code, I can immediately see things that jump out at me as reasons why this code is very reusable.  Some of them are standard OO principles, and some are kind-of home grown litmus tests: Single Responsibility Principle (SRP) – The only reason this extension method need change is if the Thread class itself changes (one responsibility). Stable Dependencies Principle (SDP) – This method only depends on classes that are more stable than it is (System.Threading.Thread), and in itself is very stable, hence other classes may safely depend on it. It is also not dependent on any business domain, and thus isn't subject to changes as the business itself changes. Open-Closed Principle (OCP) – This class is inherently closed to change. Small and Stable Problem Domain – This method only cares about System.Threading.Thread. All-or-None Usage – A user of a reusable class should want the functionality of that class, not parts of that functionality.  That’s not to say they most use every method, but they shouldn’t be using a method just to get half of its result. Cost of Reuse vs. Cost to Recreate – since this class is highly stable and minimally complex, we can offer it up for reuse very cheaply by promoting it as “ready-to-go” and already unit tested (important!) and available through a standard release cycle (very important!). Okay, all seems good there, now lets look at an entity and DAO.  I don’t know about you all, but there have been times I’ve been in organizations that get the grand idea that all DAOs and entities should be standardized and shared.  While this may work for small or static organizations, it’s near ludicrous for anything large or volatile. 1: namespace Shared.Entities 2: { 3: public class Account 4: { 5: public int Id { get; set; } 6:  7: public string Name { get; set; } 8:  9: public Address HomeAddress { get; set; } 10:  11: public int Age { get; set;} 12:  13: public DateTime LastUsed { get; set; } 14:  15: // etc, etc, etc... 16: } 17: } 18:  19: ... 20:  21: namespace Shared.DataAccess 22: { 23: public class AccountDao 24: { 25: public Account FindAccount(int id) 26: { 27: // dao logic to query and return account 28: } 29:  30: ... 31:  32: } 33: } Now to be fair, I’m not saying there doesn’t exist an organization where some entites may be extremely static and unchanging.  But at best such entities and DAOs will be problematic cases of reuse.  Let’s examine those same tests: Single Responsibility Principle (SRP) – The reasons to change for these classes will be strongly dependent on what the definition of the account is which can change over time and may have multiple influences depending on the number of systems an account can cover. Stable Dependencies Principle (SDP) – This method depends on the data model beneath itself which also is largely dependent on the business definition of an account which can be very inherently unstable. Open-Closed Principle (OCP) – This class is not really closed for modification.  Every time the account definition may change, you’d need to modify this class. Small and Stable Problem Domain – The definition of an account is inherently unstable and in fact may be very large.  What if you are designing a system that aggregates account information from several sources? All-or-None Usage – What if your view of the account encompasses data from 3 different sources but you only care about one of those sources or one piece of data?  Should you have to take the hit of looking up all the other data?  On the other hand, should you have ten different methods returning portions of data in chunks people tend to ask for?  Neither is really a great solution. Cost of Reuse vs. Cost to Recreate – DAOs are really trivial to rewrite, and unless your definition of an account is EXTREMELY stable, the cost to promote, support, and release a reusable account entity and DAO are usually far higher than the cost to recreate as needed. It’s no accident that my case for reuse was a utility class and my case for non-reuse was an entity/DAO.  In general, the smaller and more stable an abstraction is, the higher its level of reuse.  When I became the lead of the Shared Components Committee at my workplace, one of the original goals we looked at satisfying was to find (or create), version, release, and promote a shared library of common utility classes, frameworks, and data access objects.  Now, of course, many of you will point to nHibernate and Entity for the latter, but we were looking at larger, macro collections of data that span multiple data sources of varying types (databases, web services, etc). As we got deeper and deeper in the details of how to manage and release these items, it quickly became apparent that while the case for reuse was typically a slam dunk for utilities and frameworks, the data access objects just didn’t “smell” right.  We ended up having session after session of design meetings to try and find the right way to share these data access components. When someone asked me why it was taking so long to iron out the shared entities, my response was quite simple, “Reuse is hard...”  And that’s when I realized, that while reuse is an awesome goal and we should strive to make code maintainable, often times you end up creating far more work for yourself than necessary by trying to force code to be reusable that inherently isn’t. Think about classes the times you’ve worked in a company where in the design session people fight over the best way to implement a class to make it maximally reusable, extensible, and any other buzzwordable.  Then think about how quickly that design became obsolete.  Many times I set out to do a project and think, “yes, this is the best design, I can extend it easily!” only to find out the business requirements change COMPLETELY in such a way that the design is rendered invalid.  Code, in general, tends to rust and age over time.  As such, writing reusable code can often be difficult and many times ends up being a futile exercise and worse yet, sometimes makes the code harder to maintain because it obfuscates the design in the name of extensibility or reusability. So what do I think are reusable components? Generic Utility classes – these tend to be small classes that assist in a task and have no business context whatsoever. Implementation Abstraction Frameworks – home-grown frameworks that try to isolate changes to third party products you may be depending on (like writing a messaging abstraction layer for publishing/subscribing that is independent of whether you use JMS, MSMQ, etc). Simplification and Uniformity Frameworks – To some extent this is similar to an abstraction framework, but there may be one chosen provider but a development shop mandate to perform certain complex items in a certain way.  Or, perhaps to simplify and dumb-down a complex task for the average developer (such as implementing a particular development-shop’s method of encryption). And what are less reusable? Application and Business Layers – tend to fluctuate a lot as requirements change and new features are added, so tend to be an unstable dependency.  May be reused across applications but also very volatile. Entities and Data Access Layers – these tend to be tuned to the scope of the application, so reusing them can be hard unless the abstract is very stable. So what’s the big lesson?  Reuse is hard.  In fact it’s damn hard.  And much of the time I’m not convinced we should focus too hard on it. If you’re designing a utility or framework, then by all means design it for reuse.  But you most also really set down a good versioning, release, and documentation process to maximize your chances.  For anything else, design it to be maintainable and extendable, but don’t waste the effort on reusability for something that most likely will be obsolete in a year or two anyway.

    Read the article

  • Where did ULSTraceLog go to in the SharePoint 2010 Logging Database?

    - by Jan Tielens
    The Logging Database is one of the many new concepts that will make the life of many SharePoint administrators quite a bit more enjoyable. In SharePoint 2007 the Unified Logging System (ULS) logged all of its data to text files, typically found on your SharePoint server in 12\LOGS. We still have that in SharePoint 2010, but besides those text files, ULS can also write the data to a database! The advantages are obvious: easy to query, one central location for all servers in the farm, easy to build reports etc. You can find this ULS data in the SharePoint 2010 logging database (typically called WSS_Logging), in the view ULSTraceLog. Quite recently on one of my demo machines (standalone installation on Windows 7) I noticed the ULSTraceLog view was not available in the logging database. It turned out that there is a Timer Job that’s responsible for writing the data to the database, when the Timer Job hasn’t executed, the view is not there (the first time it executes, the view is created). Even more, the timer job was disabled, so the view would never be created, nor any data would be written to the database. If you encounter this situation as well, it’s quite easy to solve: Open the SharePoint Central Administration site Navigate to the Monitoring section Select Review Job Definitions Click on the job with the name Diagnostic Data Provider: Trace Log Click on the Enable button to enable it Optionally click on Run Now afterwards, to start it immediately There you go, the ULSTraceLog will be created and the ULS messages will appear in the database!

    Read the article

  • Screen lock broken when using separate x screens

    - by Wolf
    For some reason, the screen lock appears to be broken when I use separate x screens (it works fine if I use twin view or just one monitor, though). If I wait for the screen to lock (or if I tell it to lock), it fades to black, showing only the mouse cursor. However, when I move the mouse to bring up the password prompt, nothing happens, the screens remain completely blank. Furthermore, it would appear that the screen lock never activated in the first place, for two reasons: one, if I blindly enter my password and press enter, nothing happens; and two, if I press ctrl+alt+delete and then enter it logs me out (which shouldn't happen if the password prompt is there, right), losing any progress I have on any of the open programs (meaning, not a very friendly option for a programmer ;) ). So yeah, does anyone know why this is happening, and how to fix it? Thanks! Notes: I am using Ubuntu 11.10 64 bit. I am using gnome-classic, but the same problem occurs when I use unity. I do not have xscreensaver, nor do I have any screen savers running (except, I guess for the default blank one).

    Read the article

  • Unit testing statically typed functional code

    - by back2dos
    I wanted to ask you people, in which cases it makes sense to unit test statically typed functional code, as written in haskell, scala, ocaml, nemerle, f# or haXe (the last is what I am really interested in, but I wanted to tap into the knowledge of the bigger communities). I ask this because from my understanding: One aspect of unit tests is to have the specs in runnable form. However when employing a declarative style, that directly maps the formalized specs to language semantics, is it even actually possible to express the specs in runnable form in a separate way, that adds value? The more obvious aspect of unit tests is to track down errors that cannot be revealed through static analysis. Given that type safe functional code is a good tool to code extremely close to what your static analyzer understands. However a simple mistake like using x instead of y (both being coordinates) in your code cannot be covered. However such a mistake could also arise while writing the test code, so I am not sure whether its worth the effort. Unit tests do introduce redundancy, which means that when requirements change, the code implementing them and the tests covering this code must both be changed. This overhead of course is about constant, so one could argue, that it doesn't really matter. In fact, in languages like Ruby it really doesn't compared to the benefits, but given how statically typed functional programming covers a lot of the ground unit tests are intended for, it feels like it's a constant overhead one can simply reduce without penalty. From this I'd deduce that unit tests are somewhat obsolete in this programming style. Of course such a claim can only lead to religious wars, so let me boil this down to a simple question: When you use such a programming style, to which extents do you use unit tests and why (what quality is it you hope to gain for your code)? Or the other way round: do you have criteria by which you can qualify a unit of statically typed functional code as covered by the static analyzer and hence needs no unit test coverage?

    Read the article

  • Conditional attribute in XML - most concise solution?

    - by Lech Rzedzicki
    I am tasked with setting up conditional profiling - a method of tagging chunks of XML with an attribute, which will then be used as a conditional value to extract subset of that XML. Have a look at another definition/example: DITA profiling The XML is documents that are equivalent to printed books - i.e. documents that are often looked at by a human, even if indirectly. Therefore I am looking at a few requirements here: 1. keeping the value list brief - so it doesn't affect the readability of the document 2. be able to process with standard XML tools - a space-separated list inside an attribute is still probably fine, but I'd rather not use too much regexp for this 3. be obvious for various users, including 3rd parties, which content goes where 4. Be easy to maintain going forward Therefore one easy solution is: The problem with this: 1. As the list grows the value of the attribute can be a bit verbose 2. One needs to explicitly state every value even if it's a scenario of this vs everything else Therefore I am also looking at other approaches such as: 1. Using + and - modifiers, Apache htaccess style to override the default cascading of profiling - by default all content goes everywhere and if we want to exclude a bit we just say "-kindle". It does require parsing the whole tree, is not supported by editing tools and one needs to regexp the attribute value a bit deeper... 2. Using an intermediate file to define groups of values such as "other" or "non-print", example of this in DITA. It allows concise XML as well as different grouping and values for each document but it does create a certain level of abstraction which may make it a little less obvious for a 3rd party? Altogether, if you received such XML and were tasked to process it, which option you'd rather receive? If you have any experiences like that, even in an unrelated areas such a builds, don't hesitate to comment!

    Read the article

  • Some Nice Icon Themes for Ubuntu [on hold]

    - by Saurav Kumar
    Last night I was searching for some good looking icon themes availabe for Ubuntu, but unfortunately could not find any post which gives a list of themes :(. From yesterday night I started searching icons themes and installing one by one. So thought to share some of them hope it will help others to get a bunch of them in one post. This is a Q&A-Style tutorial which will cover: Few good icon themes which are used in famous Ubuntu type distros and are available in repository. Look and feel through pictures. How to easily install? I also request others to share their experiences of some awesome icon themes. Once themes are installed Ubuntu-Tweak will help to change and manage them, since I could not find another way to change the themes in Unity. In xfce(Xubuntu) it can be changed from Appearance. Just believe me this post will not be vast. If I'll remove pictures then it will become small, but it will not look good. Please give suggestion to make it small.

    Read the article

  • Which mobile operating system should I code for?

    - by samgoody
    It seems as though mobile computing has fully arrived. I would like to rewrite two of our programs for mobile devices, but am a bit lost as to which platform to target. Complicating this decision: I would need to learn the relevant languages and IDEs - my coding to date has been almost all web based (PHP, JS, Actionscript, etc. Some ASPX). Most users seem to be religious about their mobile decision, so oral conversations leave me more confused then enlightened. I do not yet own a smartphone - will have to buy one once I know which platform to be aiming for. Both of my programs are more for business users, (one is only useful for C.P.A.s). I am a single developer, and cannot develop for more than one platform at a time. Getting it right is important. Based on what I've found on the web, I would've expected RIM to be a shoo-in, and the general order to be as follows: RIM Blackberry - More of them than any other brand. Despite naysayers, they've had double the sales (or perhaps 5X the sales) of any other smartphone, and have continued to grow. And, they have business users. Android - According to Schmidt, they have outsold everyone else except RIM (though I can't find where I read that now), and they are just getting started. According to Comscore, they are already at 8% of the market and expected to hit Shcmidt's claims within six months. Nokia - The largest worldwide. If they would just make up between Maemo or Symbian, I would be far less confused. iPhone - Much more competition by other apps, fewer sales to be had, and a overlord that can delay or cancel my app at any time. Is Cocoa hard to learn? Windows Mobile - Word is that version 7 will not be backwards compatible and losing market share. Palm WebOS - Perhaps this should go first, as it is the only one that offers tools to make my life easy as a web application developer. No competition in marketplace. But not very many users either. However, a search on StackOverflow shows a hugely disproportionate number of iPhone questions versus Blackberry. Likewise, there are clearly more apps on iPhone, so it must be getting developer love. What is the one platform I should develop for? Please back up your answer with the logic.

    Read the article

  • Why can't I run Compiz?

    - by jasoncruz98
    I installed Ubuntu 11.10 on my laptop. The main reason I switched to Ubuntu is because I wanted to use Compiz. The first thing I did was to go to Additional Drivers and install ATI/AMD Proprietary FGRLX Graphics Driver. There was also another one available, ATI/AMD Proprietary FGRLX Graphics Driver (post-release updates), but I didn't install that one, because it basically meant the same thing to me as the one I already installed. Next, I went to the ubuntuguide.org Oneiric Wiki http://ubuntuguide.org/wiki/Ubuntu:Oneiric#Compiz_Fusion So I followed the instructions there and ran this command in terminal: sudo apt-get install compiz compizconfig-settings-manager compiz-fusion-plugins-main compiz-fusion-plugins-extra emerald librsvg2-common But then, the terminal window said that the package "emerald" could not be found. So, I ran this command instead: sudo apt-get install compiz compizconfig-settings-manager compiz-fusion-plugins-main compiz-fusion-plugins-extra After that, I installed Fusion Icon by running this command: sudo apt-get install fusion-icon I restarted my computer, searched for Compiz Config Settings Manager, and clicked on it. Then, I activated Wobbly Windows. I logged off and logged back on again, but there was no wobbly windows effect. So I tried clicking on Fusion Icon, but it never started. Can someone please tell me what I did wrong here? Because I see everyone seems to be able to run Compiz except me. I really need to start Compiz, or else I think I'm going to uninstall Ubuntu.

    Read the article

  • Constructor vs setter validations

    - by Jimmy
    I have the following class : public class Project { private int id; private String name; public Project(int id, String name, Date creationDate, int fps, List<String> frames) { if(name == null ){ throw new NullPointerException("Name can't be null"); } if(id == 0 ){ throw new IllegalArgumentException("id can't be zero"); } this.name = name; this.id = id; } public int getId() { return id; } public void setId(int id) { this.id = id; } public String getName() { return name; } public void setName(String name) { this.name = name; } } I have three questions: Do I use the class setters instead of setting the fields directly. One of the reason that I set it directly, is that in the code the setters are not final and they could be overridden. If the right way is to set it directly and I want to make sure that the name filed is not null always. Should I provide two checks, one in the constructor and one in the setter. I read in effective java that I should use NullPointerException for null parameters. Should I use IllegalArgumentException for other checks, like id in the example.

    Read the article

  • TechEd Europe early bird saving &ndash; register by 5th July

    - by Eric Nelson
    Another event advert alert :-) But this one comes with a cautious warning. I spoke at TechEd Europe last year. I found TechEd to be a huge, extremely well run conference filled with great speakers and passionate attendees in a top notch venue and fascinating city. As an “IT Pro” I think it is the premiere conference for Microsoft technologies in Europe. However, IMHO and those of others I trust, I didn’t think it hit the mark for developers in 2009. There was a fairly obvious reason – the PDC was scheduled to take place only a couple of weeks later which meant the “powder was being kept dry” and (IMHO) some of the best speakers on developer technologies were elsewhere. But I’m reasonably certain that this won’t be repeated this year (Err… Have I missed an announcement about “no pdc in 2010”?) Enjoy: Register for Tech·Ed Europe by 5 July and Save €500 Tech·Ed Europe returns to Berlin this November 8 – 12, for a full week of deep technical education, hands-on-learning and opportunities to connect with Microsoft and Community experts one-on-one.  Register by 5 July and receive your conference pass for only €1,395 – a €500 savings. Arrive Early and Get a Jumpstart on Technical Sessions Choose from 8 pre-conference seminars led by Microsoft and industry experts, and selected to give you a jumpstart on technical learning.  Additional fees apply.  Conference attendees receive a €100 discount.   Join the Tech·Ed Europe Email List for Event Updates Get the latest event news before the event, and find out more about what’s happening onsite.  Join the Tech·Ed Europe email list today!

    Read the article

< Previous Page | 407 408 409 410 411 412 413 414 415 416 417 418  | Next Page >