NetworkedPlanet Blog » Uncategorized

Concepts for EPiServer 1.1.4 Released

Andrew Turley — Fri, 09 Dec 2011 11:58:22 +0000

Concepts for EPiServer 1.1.4 is now available for download.

This version fixes a specific issue where the module would prevent page saving when the module is not yet configured for a topic map. The module now ignores PageSaved events unless there is a valid topic map configured that can be connected to.

Using PowerShell To POST Data to Web3

Kal — Wed, 18 Aug 2010 11:01:19 +0000

Sometimes it is useful to be able to use PowerShell to manipulate a Web3 instance - either because you are scripting a set of batch operations or because you are just in PS at the time and it saves swapping contexts.

POSTing data to a REST service with PS is really easy, the only small wrinkle with Web3 is the need to authenticate with the server. Assuming you are using the default Windows Authentication setup for your Web3 server, the following set of commands should get the job done:

$wc=New-Object System.Net.WebClient
$cred=Get-Credential
$wc.Credentials=$cred
$wc.UploadString($URI, $CONTENT)

The Get-Credential cmdlet will pop-up a dialog prompting for your username and password - if anyone knows of a way to avoid that please let me know! If you need to load the content from a file, use the Get-Content cmdlet.

Enterprise Services 1.1 Hotfix 2 released

Jen — Thu, 10 Dec 2009 14:21:16 +0000

We have just released a new version of Enterprise Services available for download. This release changes the way topics passed in from a result set are processed so that the search results retain the ordering of the topics passed in.
Please check the instructions contained in UPGRADE.TXT for important information on the upgrade process.
Download the updated package .

Implementing Topic Maps Content Management Integration using Atom and Live Mesh

Graham Moore — Tue, 28 Jul 2009 13:12:13 +0000

One of the most common scenarios for Topic Maps is to use them as the primary paradigm for organising, classifying and finding content in a content management system. The basic strategy involved in these integrations is to create topics for each content item in the CMS and then associate the content topics to other topics in the the topic map. We call this process 'content item proxification and classification'. This builds up a nice graph that can then be exploited in search and navigation controls either within the CMS or as part of some other publishing process.
The key to making this work is to somehow hook into what is happening within the CMS so that the topic map view of the content items correctly reflects things in the CMS. i.e. if a document is deleted its topic shouldn't show up in the topicmap.
The integrations we have done for EPiServer and SharePoint have taken the approach of event handlers installed into the CMS product. This means that as things happen in the CMS our handlers are notified and can ensure that the topic map is in a consistent state.
However, more and more, we are looking at online content systems, and how to integrate these content stores with Topic Maps systems.The following sections discuss a generic pull approach to synchronisation and then give some detailed insights into implementing this approach with Live Mesh. Live Mesh is an online service from Microsoft for collborating, file syncing and generally supporting dynamic ad-hoc work groups. It is of course missing a powerful metadata management and classification aspect.
OK, so first up is the general notion of a pull based approach to integrating a Topic Maps system with a CMS. This approach was inspired by the TMShare protocol that was developed at CEN. This was an Atom based protocol that published a feed of changes that had occurred inside the system. A client could process this feed to get the current state of each item that had changed. Note: this approach didn't say what had changed, just that a given topic has changed and here was the address of its representation. The client could then update its local store to be aligned with the 'server' node.
So, instead of syncing between two topic maps this approach looks to sync the Atom feed from Live Mesh into a topicmap.
The Live Mesh environment consists of an online virtual folder and file hierarchy and optionally corresponding syncronised folders on multiple devices. However, when synced all the 'actions' performed can be found in an Atom feed for each users' Mesh. The news feed for a Mesh can be found at the following URL (note that the base of the URL will change as this is the CTP):


https://user-ctp.windows.net/V0.1/Mesh/News

This feed contains a series of entries. Each entry corresponds to some event that has occurred; such as a file being added or a folder deleted, renamed etc.
Our syncronising application runs as a background process on the users' machine. The process will fetch the feed of events and then update the topicmap accordingly (this topic map could be local or in some online Topic Maps service). In a local TMCore instance I have already set up an ontology consisting of File, Folder, Mesh Space, Concept and Person. The items in Mesh will map to their obvious counterparts in this ontology.
The TMShare protocol provides links to representations of things that have changed. The client is expected to pull the representation and update its local version. Mesh has a more transaction based approach where is lists the actions that have been performed. However, if a client application just sticks to doing changes rather than updating full representations then there is no way for a client to get back in sync if an event is lost or if some events simply are not published (except for an out of bound action). For this integration we consider each event to simply be a notification that a given Mesh entity has been updated / deleted. We then fetch that entity's Atom entry and from that update the topic representation. For most content management integrations we can accept some delay in getting in sync but we would like it to happen.
Drivin the sync process is the News feed. It contains an entry for each event that has occurred. Here is the basic structure of a 'Live Mesh Event Atom Entry':
urn:uuid:2979996d-57a0-e3c7-d7bd-b5f1a2b33bcb LiveMesh.LiveFolderAdd 2009-07-28T12:43:35Z 2009-07-28T12:43:35Z Graham Moore email-graham_moore@live.com/Profiles graham_moore@live.com 1 Scope LiveFX/MeshObject Mesh/MeshObjects/3XDCEOEAANZUPFD5OV7TLE3IFM documents41 LiveFX/MeshObject Target LiveFX/MeshObject Mesh/MeshObjects/3XDCEOEAANZUPFD5OV7TLE3IFM documents41 LiveFX/MeshObject
The key things that we use from the above Atom entry is the Title element value. This is the unique key for the event type. I would have preferred to see this implemented a little more robustly, either in a custom attribute/element in the entry or as part of the content. I certainly would have liked to see it as a URI rather than just a string.
The following are the list of event types that can occur in the Live Mesh Environment:
LiveMesh.FolderRename - Occurs when a folder is renamed (note there is no event created when a LiveFolder is renamed.) LiveMesh.FolderAdd - Occurs when a new folder is added. LiveMesh.MemberAdd - Occurs when a member is added to Folder. LiveMesh.LiveFolderAdd - Occurs when a new LiveFolder is added. Why the need for the special cases? If the scope was set to the Mesh that would be enough. LiveMesh.LiveFolderDelete - Occurs when a LiveFolder is deleted. Note: no events are raised for all the sub files and folders that are also removed. LiveMesh.FileAdd - Occurs when a file is added to a folder. LiveMesh.FileUpdate - Occurs when a file is updated or renamed. LiveMesh.FileDelete - Occurs when a file is deleted. LiveMesh.MemberInvite - Occurs when a member is invited to join a folder.
With the event detected we can then pull information out of the Contexts element. This data structure isn't the cleanest for pulling out the key information but it tells us on which item the event has occurred and in what context (folder). The NewsContext provides links to the items being affected. These items can then be fetched and the state replicated into the topicmap. This includes creating and deleting topics for files and folders, maintaing relationships between files and folders and folders and their parents.
As Live Mesh doesnt raise events for recursive deletes, we fire a pseudo delete event when there is a LiveMesh.LiveFolderDelete event for all descendent files and folders.
All of this processing means we then have a topicmap synced with the state in a given Live Mesh account.
The trade off we see between an event based integration and a pull integration is that the pull integration is more lightweight, robust at dealing with failure, and easier to evolve over time. The value of the integrated event model based approach is that information moves in near real time from the CMS into the Topic Maps system.
So by using the Atom feed of changes to a Live Mesh environment we are able to create and maintain proxy topics for all files and folders. This gives us a baseline on which to build further functionality such as advanced metadata management, classification, search etc.
Now all we need is a nice Silverlight UI to allow users to navigate, query and further classify all their nice Mesh file topics

Article (in german) about the TM based EGovShare Protocol

Graham Moore — Thu, 04 Jun 2009 08:41:47 +0000

Marc Kuster pointed me at a very positive article about the EGovShare protocol developed as part of a CEN work activity.
The article is
http://www.egovernment-computing.de/projekte/articles/188135/
And, Marc commented that ''you might find this little article intersting that was published without direct intervention from me. Especially the statement from our federal ministry of the interior is astonishingly positive."

Email Outage

Kal — Fri, 12 Dec 2008 12:40:39 +0000

Actually there should be no outage, we hope...but its always best to plan for the worst
This weekend (12th-15th December) we will be migrating our mail servers which means there may be a short period of outage. This will mainly affect the web form for requesting product evaluation licenses (which will be disabled for the period of the move) and may also result in messages to support@networkedplanet.com not receiving the normal automated reply.
The plan is that all email should still be delivered to us, but if you send us an email over this weekend and don't get a reply by Monday please try sending it again.

Opening downloaded help (.chm) files

andy — Tue, 09 Jan 2007 16:27:14 +0000

We have noticed that when downloading the TMCore EPiServer Module it is sometimes the case that the help file, .chm file does not open correctly. This article explains why, and how to fix it.

The solution

Rather than tantalize and tease with the solution, to make the chm file readable on your machine open up the properties for the CHM file and click the Unblock button as shown here below:

The problem

The "problem" is caused by Windows protecting your system. Windows is aware that the CHM file originated from the internet. The help system, which is HTML based, contains code (ActiveX) which, if maliciously written, could damage your machine. Therefore, Microsoft issues are security patch (896358/890175) to disable this by default.

In order to allow the code within the CHM file to run, you must "unblock" the file.

References

The links below explain the exact nature of the problem and alternative workarounds to this problem.