Picture
Well fellow SecondLife residents, I wanted to give you all an update about my idea of introducing Virtual Landmarks in SL.  Since my first introductory blog posting on July 24th 2012 explaining what VLMs are and why I felt so passionately that SecondLife badly needed this solution, a lot has happened.

I also have very exciting news to share about the progress of VLMs to becoming a reality in SL, but let me first catch you up.  Good news at the very end :)


FIRST - Introduce to the SL Forum Community for feedback:
After I posted my blog, I also introduced the VLM solution in both the SL Merchant & Server forums to solicit feedback from the extremely knowledgeable and skills talents that participate in these forums.  This would be the first critical hurdle for the VLM idea to jump if there was ever any hope that VLMs would see the light of day as a solution on the SL grid.  If these wise SL residents gave VLMs a negative or even mixed feelings then VLMs would be dead right out of the gate.  There is little chance for LL to listen to most SL resident ideas at the best of times but there would be ZERO chance if the community was not 100% strongly liking the need for the idea.

I presented the idea i both Server and Merchant forums because each forum has a different population of experienced SL residents. I needed to get a big thumbs up from both.  The great news was that the overwhelming response about VLMs was "THIS IS AN AWESOME BADLY NEEDED IDEA AND ITS EXTREMELY WORKABLE AS A SOLUTION AND FLEXIBLE TO BE EXPANDED FOR OTHER PURPOSES".  

The only negative that came from these forum communities was the idea of LL charging a possible use fee to create VLMs.  That was only a side option to the technical solution so that was easy to address since it was not mandatory.

Another major purpose for posting a thread in these two forums was that they could help fine-tune and point out possible improvements or corrections to the overall VLM solution I initially posted.  This proved valuable as a few things were improved and corrected.  Most notable was the description and term of the central service.  I initially characterized this service very similar to a DNS in that the focus was providing "unique LM names" but that was not actually the case.  The visible name of the LMs in the service were irrelevant to the solution.  It was actually a VLM "MAPPING SERVICE".  So the revised design shows the solution having a VLM-MS.


SECOND - Develop formal SL JIRA to introduce it as a NEW FEATURE to LL:
Now that I had a solution that a critical "bell weather" sampling of the SL resident community both liked and helped me revise and perfect, it was time for me to formally introduce the concept and design of the VLM solution to LL.  Until this week, introducing a new proposed feature to LL was accomplished through SL's JIRA system.  That is what I did.  


Sadly this week Rodvik (LL's beloved CEO) decided to shut down the SL Resident's JIRA so as of this blog I cannot provide you the link to this JIRA that was submitted - that is another blog topic I posted yesterday.

Regardless, I formally laid out the technical aspects of the VLM solution for LL and the rest of the SL Resident community to understand, VOTE, and comment on.  The formal JIRA also included architecture diagrams of how the solution works now and how it would work with the VLM-MS in place.  You can see the two diagrams below - side-by-side.

Picture
The current LM solution - without VLMs (click to enlarge)
Picture
How LMs would work with VLMs in place (click to enlarge)
It included all the details of the revised design of using a VLM-MS (Mapping Service) and also provided a list of the several potential benefits to the solution.


THIRD - Socialize VLM Solution to SL Community / Seek Strong Support:
The next step was the most difficult and is where 99.9% of any good idea falls dead never to be heard from again - especially when the idea needs to be adopted by LL.  The primary objective in this phase was to "GET LL TO SEE THE IDEA AND TAKE THE TIME TO REALLY UNDERSTAND IT".  To do this, it was critical to first socialize it openly into the greater SL resident community.  I did this by promoting my initial blog and more importantly the new VLM JIRA in all the social media tools I currently had established.  I need to get the community to go read the JIRA and vote for the JIRA and even post positive and technical inputs / opinions to the idea.  I posted on Twitter, Facebook, SLUniversity, the SL forums, and reached out to all SL connections who also have community reach in hopes to make the idea become viral.
It was also critical to generate a BUZZ WORD that the community could attach to when talking about and spreading the word about the idea.  "VLMs" was it.  Basically it is like creating a BRANDNAME.  Without it, the message would be hard to migrate to viral status if everyone called it their version of "that idea from Toy".

To further strengthen the marketing of the idea, I needed a marketing poster/logo to promote the VLM brand.  I used my artistic talents to create a VLM Logo as you can see to the right.  I posted the VLM logo everywhere and I created a VLM Poster inworld with info inside and offered it to the Merchant community to get a copy and have them post it in their stores, malls, art galleries, clubs.

The results were great as I got a lot of positive buzz from all the community.  And although you can no longer see the JIRA (thanks to Rodvik's dumb JIRA GAGGING this week) the JIRA reached 96 votes & positive input in 2-3 weeks !  
Picture
Sadly the JIRA is no longer visible to SL Residents
FOURTH -  Personally Present VLM JIRA to LL Developers -  Aug 3rd 2012:
So with all my "ducks in a row", the next critical step was to get LL Developers to take notice of the VLM JIRA.  As we all know so many JIRAs are never looked at by LL staff.  I had to get it in their face where they could not hide and pretend they didn't hear it.  I was told to present the idea at an up-coming SecondLife User Group meeting.  These obscure weekly meeting are hosted by LL developers / tech staff and ideas / problems are openly talked about.  After some talks with fellow residents, the decision was made to introduce it to the Server User Group meeting on Friday August 3rd at 4pm SLT.

To add strength to the VLM JIRA presentation, I could not go there alone to present the idea.  Firstly because I don't think many LL staff really like me as I am often critical of many of their decisions;  Secondly, because I needed champion voices at the meeting to both endorse the VLM idea and add their logic opinions on how important the idea is to the merchants / club / art gallery community.  Sadly, LL staff run SL but don't know how their customers actually use SL and the pains they must suffer with within SL.   As such, I mentioned to the merchant community of my plan to present the idea to LL at the meeting and asked anyone interested to come provide your presence and support at the meeting.  Several Merchants attended with me.

At the UG meeting I formally introduced the VLM JIRA to the group including Andrew, Simon, Cheesey, & Baker Linden.  The presentation went very well.  I was also extremely impressed at how much healthy discussion was had for over 20 minutes about the VLM idea.  The Lindens quickly grasped the design behind it and saw a lot of potential in the design - not only for its immediate benefits but how the service could be expanded to provide other similar functions from a central position.  You can see the entire discussion yourself as these UG meeting transcripts are posted:

http://wiki.secondlife.com/wiki/Simulator_User_Group/Transcripts/2012.08.03

An extremely value side-benefit of the meeting was that one of the SL Resident attendees - Nalates Urriah - is a very popular and well followed blogger of all these technical LL UG meetings.  After the meeting she dramatically helped expand the socialization of the VLM solution to a far greater SL community when she created an excellent objective blog posting on VLMs.  I have provided a link to her blog posting of VLMs as it was very well written.

http://blog.nalates.net/2012/08/04/virtual-landmarks/

Since then, there have been other excellent blogger postings of it.


FIFTH - Keep LL's Feet To The Fire on VLMs & Don't Let VLMs Become Forgotten:
As is the case with most good ideas, the excitement of the idea starts to fade and becomes forgotten.  If it gets forgotten by the community then its easy for LL to sweep it away as a "good idea but no time".  I couldn't control the excitement fading, but what I could do is attend every subsequent SL Server UG Friday meeting and ask the Lindens for a status update.

I went every week and generally got the message from them that "LL is reviewing the VLM idea for its viability and design implementation risks and we will keep you posted".  I expected this kind of answer or the "nothing has changed" answer each week as I attended, but I felt it was important to keep showing them that I wasn't going to let VLM's become forgotten.


FINALLY - THE GOOD NEWS!!  PROGRESS !!
So this is what you all have been waiting for  (unless you cheated and just scrolled down to there right away ).

At the Friday September 7th, 2012 UG meeting that I attended, I asked the same question that I asked in all the other past meetings.  BUT - this time I got a different answer from Cheesey and Andrew that is very exciting for VLMs.  First let me paraphrase and then show you the actual transcripts messages:

Basically, LL has completed their feasibility review of the VLM JIRA and deemed it a viable solution -  one that warrants them proceeding to the phase of setting it within their development priorities!

Here is the link to the actual transcript and my clipped out portion of the topic:

http://wiki.secondlife.com/wiki/Simulator_User_Group/Transcripts/2012.09.07

[16:19] Toysoldier Thor: and Cheesey - and progress on assessment on the VLM jira?
[16:19] Cheesey Linden: Toy it's been vetted and is not (meant to say "now") in products hands to prioritze
[16:20] Toysoldier Thor: is that good news Cheesey?
[16:20] Cheesey Linden: yes.
[16:20] Cheesey Linden: that is progress
[16:21] Andrew Linden: I think he meant to say: "... it is *now* in product's hands..." where "product" is our word for an internal team
[16:21] Cheesey Linden: woops yes i meant now
[16:21] Toysoldier Thor: yup i picked that up - thanks Andrew

After the meeting I talked to another Linden and asked his/her's honest opinion if VLMs really have a shot of becoming a developed "priority" or if I was just wasting my time promoting.  I am summarizing and paraphrasing our chat but basically "good news" feedback came back.

The Linden said that internally, the VLM idea got far more traction than he/she expected it would (with so many other priorities on the LL books) so there is hope for it.  He/she said that "the product guys got it right away"  and there were other voices internally that were saying how the VLM solution would help other areas for LL as well.

He/she said the only aspect that seemed to concern the internal teams (called it a technical problem to solve) was the VLM-MS service's ability to handle the VLM request loads.  I told him/her that this is by far not a problem and one of the easiest things to address.  

The queries to the VLM-MS are extremely simple and static requests to these servers.  I mentioned how DNS servers easily handle several times more query loads than this.  And that even an Active Directory Domain controller likely would get hit many more times for AD and LDAP calls than this service would.  I also said that because of the structure of the VLM-MS, these servers could esily be placed behind a load balancer and/or a proxy cache could be placed in front, and/or deployment of caches could be thought of (although I do not think that would be needed).  He/she agreed.


CONCLUSION - GREAT PRGRESS - MORE TO GO !
So that is where we stand now and that is great news - even though VLMs still have a long way to go before we all will be handing out VLMs to each other.  

Keep promoting the idea of VLMs..... Keep mentioning their value to any Linden you meet on the street.



Leave a Reply.