Few counter points:
· We are talking past each other, i was actually on the defense given the assumption i thought you were projecting was to NOT use a plug-in at all for a public facing website and stick to using HTML/JS solutions. I agree, hybrid solutions for public consuming solutions are actually a recommendation I've given countless times. There is no need to make 100% driven plug-in experience for a site, given its utterly failed every attempt in the past - MTV, EA GAMES etc. · However, if you spend X number of dollars for demand gen and associate having a plug-in experience only is considered an automatic waste of money, I'd simply say you're wrong. As despite sites like MTV/EA GAMES etc doing such, they still retained high numbers in traffic and to even put the smack down on that argument actually retained a higher stickiness then their HTML counter parts. End users at times switch gears and treat a site like a form of entertainment etc rather than a typical site, it's the interruption factor that yields positive returns here. I simply think generalizing that plug-in experience = High Risk is misguided is all. It always comes down to execution (both concept and talent producing it). · How do you define routinely kill? is it a case of not installing and walking away? what kind of A/B testing did you conduct? how do you even measure the level of friction associated to a "Install" gate? abandonment rates aren't an exact science, all they tell you is thumbs up thumbs down, if anything abandonment rates give you an insight into people who were *actually* interested in what it was you were soliciting them to see. People walking away isn't a number I'd argue as being one you declare success/failure on, as there are variables associated to "WHY" they came in the first place. I watched a Microsoft site get millions of hits a month and everyone high fiving one another about how popular it was, I did a closer inspection and after 1 click the number went down from millions to hundreds... 1 click... point is, it's a holistic analytical discussion that needs to occur. Throwing a plug-in under the bus is easy as it plays into mob mentality in many ways "Get em, there the reason we failed"... when you ask further questions like "What was your solicitation like? what was your selling points to install? what parts did they abandon the install process? out of the people who did install, what parts of the experience did they sit and dwell on? where did the people come from before they arrived at the site? what was the expectations of the said people?" etc... eMarketing 101 stuff. NBC Olympics and countless other sites have proven people will install a plug-in if the content is worth it. Care factor on the size and install process is THE most common feedback in situations like this. · Flash's 100% ubiquity is a marketing sucker punch. Firstly how do you go from having 99% ubiquity on a plug-in when there are 8-18 million people on earth every day installing the runtime? Do the math, there are 1.4billion people online (approx estimates - including people behind firewalls), So that's 8-18million x 365 / 1.4billion? Hmmmm..... Then you look closer at the methodology used by Adobe, and you start poking holes in it, like "Oh really, a sample of 1000 for the US but only 400 for China? really? last time i counted there were approx 380million people in the US population, let's look at China/India...oh wait..dang" - let's not forget that the samples are volunteer based and only for people 18+ to undertake (I'll let MySpace/Gaming studios know etc they are wasting their time with the below 18 market)... The reality behind Flash is not its runtime ubiquity, it's the actual saturation levels of FLASH online. You can't go an hour without being prompted to install Flash should you not have it, now this is the kicker, this goes for QuickTime and a whole bunch of other 3rd party after market OS add-ons. It's a common tax being paid daily, so in a sense Flash's constant usage conditions our consumers into the notion that installing plug-ins are a tax often paid. Disable Flash, spend an hour browsing the web, see how far you get. Ubiquity = fool's errand, Saturation = Goldmine. · Your assumptions about my background are just that, assumptions - I did have a life/career before Microsoft and it was heavily suited towards Adobe technology not to mention JS/HTML. I remember when people used to argue "JavaScript is friction" when it came to DHTML, in that I used to get punished about stupidly wanting to use XmlHttpRequest on public facing sites "Oh no, people don't like JavaScript" was a common argument. Fast forward today? what do we have running around. I built a lot of ecommerce sites for Tourism Queensland, Sunlover Holidays etc, tourism industry which had not just KPI's but ones that were watched painfully up close. I've done contract for companies like eBay and so on, I'm not talking through my butt on what i think RIA's all about, I've been doing this before RIA was even coined as a term in 2002 by Jeremy Allaire. Your arguments are weak, you're asking me what tell customers in the below hypothetical scenarios and the answer is "do what they need, not what you want"..Why use Silverlight/FLash for those scenarios if the customers are locked down behind firewalls in set SOE's. You're already behind the 8-ball before you even start the project and it's not the fault of the plug-ins, its merely the SOE being the constraint? - work with the constraints and overcome them is your answer. · Making sites work for ALL - Good luck with that exercise. As you're going to compromise on either quality for quantity or quantity for quality, you attempting to balance it out isn't going to work out unless you have a decent team + time + budget. Three factors that get left off the table when it comes to righteous soap boxing about the "Good of the web". Someone has to pay, someone has to raise the bar of expectations and interrupt the consumer in a way they aren't used to. We typically live in a 7 second attention span economy, interruption is really the key - technology is just the vehicle of delivery. Apple iPhone is only 21million units young, even on their best year they still haven't' scratched the surface of saturation levels for mobile devices, yet, they are prolific and highly visible device. Arguments like "make sure you make for 100% " is ludicrous, as you'll never win as competitors will always sacrifice the minority in order to recover the majority. From: [email protected] [mailto:[email protected]] On Behalf Of David Connors Sent: Wednesday, December 02, 2009 3:50 PM To: ozSilverlight Subject: Re: Our new silverlight site 2009/12/2 Scott Barnes <[email protected]<mailto:[email protected]>> Disagree with you on most of your points. You appear to basing a lot of assumptions on bad data such as "googleing" what the other guy does. Not really. I'm merely stating the facts as I see them and arguing that using Silverlight or other RIA platform to enhance a site should not preclude making that site accessible via standard HTML-based means for a variety of reasons. - Just because you spend x dollars to generate demand for a given site, does not guarantee a high retention of specific consumer activities. Of course not. I don't believe I said that. However, spending x dollars to do demand gen for people who can't access the site is going to lend to dissatisfaction and a waste of money. I don't see why it is so difficult for you to get your head around that there needs to be a considered balance in the use of RIA. a) Some sites absolutely require it (games, youtube, etc). b) Some sites might benefit from it, some not. Unless you're building something in category a above, I would not be mandating it. [...] - Implying an install experience increases friction is a misguided as well, as it will vary depending on context. Fair enough. We'll have to agree to disagree on that. I've watched far less friction than a runtime install routinely kill ~$25K ad flight spends. We have built and manage applications for customers that have turned over $10M in their multi-year life time so I have half a clue in this regard at least. JavaScript can only lead you so far in terms of generating an appropriate level experience, where as a plug-in is an extension of this metaphor beyond what the current browser models can carry out today. Example, Nike, Audi etc all invest heavily in plug-in driven experiences, that don't degrade to a JavaScript based only solution. The question for you rather than pound away at this point is - why? why do they do that as surely this an implied friction based of your argument below? e.g.: See http://www.nike.com/nikefootball/home/?locale=en_US&sitesrc=uslp without Flash installed. That's a business decision for them to make HOWEVER Flash has a market share is rather a different picture: http://www.adobe.com/products/player_census/flashplayer/version_penetration.html It is sufficiently close to 100% to be 100% if we're talking about v8 and above, which is not to say I think Flash is better than SL - just that your comparison is pointless (appeal to authority - just because a large shoe company you admire for software architecture advice does it doesn't make it right). - I'm also quite curious as to how you define your ROI assessment. In that how you determine what "friction" is defined as when it comes to putting a solution like Silverlight in front of your customers? I'm curious as to what rationale you use here? Is it the notion that "installation = abandonment?" Do you have any past experience or analytical data to support your claims or is this just from the hip? Groan. Yeah. We have two customer accounts where we are held accountable for incomplete transactions. What do I say to our vendor manager(s) when (s)he is chewed out by the FC? Oh yeah, those 250 course modules you sold are inaccessible because the military/edu/govt department was greeted with an "Install XYZ" dialogue when they tried to access their pre-paid courseware? What do I say when someone is trying to punch through a million bucks worth of event regos? "I'm sorry, in the interests of customer experience, no one from education, the public sector, banking and finance are coming". Clearly there is benefit in providing RIA type experiences and there are many sites out there that exist because of it. Youtube being probably the most popular example, and I cannot count the number of hours I've wasted on Kongregate. I was delighted to see a few Linux people I work with regularly actually silent when I showed them the SL smooth streaming stuff Jorke's set up for DPE (including the new iPhone support in IIS7 media services). One could argue that anything from Microsoft that can actually make a Linux person stop bashing MS for 60 seconds and actually admit it is good is indeed good technology. ... but I suspect you have next to zero experience in end to end responsibility for online applications with $ or KPI attached. At the end of the day I suspect you are incapable of appreciating any view point other than your own and/or are being argumentative for the sake of it. Your priorities should be (as others have said + especially so when there are $ at stake): 1. Make the site WORK and TRANSACT for ALL users for reasons of commerce (and anti-discrimination legality as mentioned elsewhere). 2. Enhance it, by all means, using SL or whatever else floats your boat. Again, SL does some really good stuff which is not possible by other means (on the web). The adaptive streaming is awesome. JO did (we hosted it) a bit of a world's first doing adaptive streaming over Akamai for the HammerTime FB/Twitter XBOX launch and it was outstanding. We had people all over Australia getting 1.5mbps streams or better - the quality full screen on my 24inch monitor was excellent. But all of that goodness to one side, to suggest that there is no drop out from installation of a client runtime on a site and have no fallback plan is ... insane. Anyway, this is going nowhere. I'm out. David. * By "we" I mean JO and not us. :) -- David Connors ([email protected]<mailto:[email protected]>) Software Engineer Codify Pty Ltd - www.codify.com<http://www.codify.com> Phone: +61 (7) 3210 6268 | Facsimile: +61 (7) 3210 6269 | Mobile: +61 417 189 363 V-Card: https://www.codify.com/cards/davidconnors Address Info: https://www.codify.com/contact
_______________________________________________ ozsilverlight mailing list [email protected] http://prdlxvm0001.codify.net/mailman/listinfo/ozsilverlight
