Conversation Agent: The Light Before the Storm

Indeed, and I think many see the positives of building a community as a way to boost revenue, and are rather happy go lucky on the potential downsides. Its not for the faint of heart, and it does take full blown commitment, as stuff will hit the fan at one point or another. Without a game plan in place, or at a minimum, a loose framework, things can spiral down pretty fast.

The worst case response scenario is the all too common typical happy happy face we can do nothing wrong pr speak. That almost always tends to inflame a situation and make what might have been managable albeit uncomfortable deal end up as a major crash and burn. Yet, the happy happy speak is so ingrained in the corporate culture, to upend it and get real in the face of adversity is anathema. Granted, there are many mitigating factors, by getting too real, one can run into anti-trust, SarOx, or other legal headaches, to say nothing of liability issues. In those situations it can end up being a loose loose deal all around, with the exception of council who smiles all the way to the bank irrespective of the outcome.

As a ex-community manager, and having to deal with thousands of really ticked off folks, it can be a real frustrating, and evening frightening deal at times. Heather, a CM over at flickr describes the role quite well, “its like being a pinata, people beat you with sticks and expect you to give them candy.” I think she really hit it on the mark with the following advice.

When things to go wrong, own it. Admit it. Don’t try to whitewash it. 

Granted, there were days when I spun things, sometimes, its an inevitable situation for the previously mentioned reasons. Yet, even when working the spin, if one is upfront that yes indeed it is spin, and you cant go further into it, but even that amount of transparency does lead to at least some good will. Doing so, may lead to only a small number of folks entering supervillain mode if any. Yep, been there done that too, the FBI is still on my cells speed dial. Things can get out of hand, and if need be, there are times and places to call in the feds, just make sure your privacy policy gives an out for law enforcement, or one can really hose up trust in a hurry. 

I do find it so very interesting how twitter is abuzz with all of this community stuff, and yet so few seem willing to talk about the potential hazards. Sure, some of it is proprietary for good reason. No one really likes to fess up that with user generated content, occasionally a terrorist will pop up, or that a contributor had things go so far south they are suicidal. Yet, these things can and do happen. The key is having the ducks in a row upfront, such that when bad things happen, there is a game plan in place ahead of time. And yes, that game plan also includes ones staff… a suicidal community member whether successful or not can inflict pretty severe trauma on the part of staffers if they are not prepared ahead of time. Even with preparation, its not so simple.

Over at Influential Marketing Blog, Rohit Bhargava presents 6 reasons why virtual events are becoming more popular. A couple I take issue with, but the rest are right on the mark.

His first reason is budgetary, and he is right on the mark with that, although I think the situation is more bleak than he presents, albeit it is industry independent. Ie, for some industries, conferences are an intergral part of their business, for others, they are beneficial for sure, but the funding priority ends up being pretty low. Back when I was a corp guy, I tried to make sure my team got out 2-4 times a year, and I was super tight with funds, but I saw the value in doing so. On the other hand, when budgets got slashed, and it was a call of whether to get a new piece of equipment, or travel, the equipment budget always came out on top.

The fantasy team aspect is a pretty good one, esp as one can time shift in the virtual domain, where as in person, it can be too much. Ie, if one pulls together a fantasy team for a in person analog conference, it often results in information overload in a huge way. A virtual event would provide for the ability to timeshift, esp since most virtual events archive the content. I remember the Bipolar / BiCMOS conferences from years back… it was a time to get brain fried in a huge way, as effectively it was 3 days of a firehose datadump. A virtual event on the other hand would allow one to self moderate such that info overload headaches would be minimized. Granted if one is a BiCMOS process dude, and its their lifeblood, a fantasy team for such a conference is likely pretty cool, as contrasted with a scientist working on the periphery. Based upon the hundreds of conferences I’ve attended over the years, in most cases, I was on the periphery, rather than it being a focus. (Radiation thermometry is too small a field to support an exclusive conference, even though it transcends a multitude of arenas).

The multitasking issue is where I have a problem, especially for smaller companies. Many conferences provide for total immersion in a subject arena away from the fires of the day. More than a few times, I felt the value of attending was the single subject focus, even more so than the presentations in and of themselves. Now, some might ascribe this as an issue of time management, and they are partially correct. The difficulty is, when a production line crashes, or a customer has a problem, being right there pretty much mandates stepping in, where as if one is offsite, either A, it can wait, or B, folks will solve it on their own. Of course there are the situations when a major line goes down, onsite attempts at a restart are not working, and it ends up one of those hundreds of dollars an hour deals or more. In those cases, the virtual conference would be a real saver, contrasted with last minute super spendy redeye flights back to say nothing of the headaches involved with trying to get up to speed on an issue while on a plane.

I agree with him as concerns small business as concerns the finance pov. Way back when we started out with 8 folks in 89, we made it a big point to hit conferences, as the ROI for the time and money invested was amazing, despite the fact I was Mr cheapskate. As time went on, we hit the 25-30 person management plateau, we cut back for a bit… but once past it, the doors were once again open. As far as intimidation goes… when I struck out on my own fulltime in 2002, I hit conferences the same as way as when I worked for a 2000 plus person employer. The disadvantage of course was that I had about zero buying power and company name recognition, vs the 5 figure signoff capability I had as a corp guy. I never saw these disadvantages as intimidating, but more so just a practical aspect of being new and small… of course I still had my corp contacts, and that opened a lot of doors.

Below is the old intro… looks like I blew it in a huge way LOL, good thing I am out of it

Todays will likely be a wild ride for the stock market. I told my wife to expect the Dow to drop 840 at some point, albeit it is likely there will be a bounce along the way. For some odd reason, probably just luck, I seem to be predicting the market really well… too bad I dont have money in it, but thats another story entirely. 

For those of us in manufacturing, and or physical hardware, the short term in all honesty doesnt look good. Manufacturing is very much dependent upon lines of credit, as no one pays their tab in 30 days. with many pushing the limits to the point of only paying in order to get off credit hold only for the most mission critical order. Hardware design at least on the contract side of things is quite similar. Even MSC, a key provider of MRO equipment for manufacturing plants had a ton of bad news in their conference call a couple weeks back.

Anyhow, its pretty easy to look at things from a doom and gloom pov, but we have a ton of positives that most folks in history could only dream of. The cost of entry into an internet business has never been lower, and with the ways things are going, it may well be that highly skilled US talent will be available at a fraction of the cost, to say nothing of big decreases in the offerings of offshore firms. I dont know that the tech sector will see massive layoffs, but it is exceedingly likely as prices go up, budget moonlighting will as well.

As people cut back, they tend to spend more and more time online… its interesting to note eBay is reporting significant success with the addition of banner ads, skype, and non-core business, plus the total number of listings is truely amazing for the month of October. Granted, such doesnt sell goods. but it does seem to indicate a significant increase in online advertising. despite the down turn. My guess is, as companies pair down their marketing budgets, they are looking for lower and lower costs, and online is a fraction of the cost of traditional media.

In addition, with the advent of cloud computing, albeit not without some rough starts, the problem of buildig a massiver server farm to anticipate demand has been reduced significantly.

The end result, and this is just in one narrow arena, its likely never been a better time to bootstrap an internet startup. The key of course is to get revenue positive from the get go, or at least as close to it as possible. I’ve seen it done, actually just a couple months ago. I passed off an idea to a friend, she ran with it, and sure enough, its cash flowing (and not just $5/month over expenses), so the oppurtunities are there. I almost wish I had run with it LOL. On the other hand, she has done a lot of free coding for me, so unless she becomes a multimillionaire off of it, and doesnt toss some my way  I’m happy for her.

There are a multitude of other arenas with potential, the Iphone and Gphone stuff I’ve blogged about is one of many. Even the creative folks over at Etsy seem to be on a roll. There is a lot of great oppurtunity for those with ideas, even if they lack some in cash flow to get rolling. I expect a lot of midnight oil will be burned by a ton of folks as we dig our way out of this mess.

I got an email from the Andy at commentluv this morning, and they are having a contest, but also will be opening up ad space, and that got me thinking… and then thinking some more.

There seems to be a growing trend toward the 125×125 pixel ads, and I thought cool, I have a 125X125 logo graphic, and could always use more traffic. The thing is… its a 125X125 graphic, its not an ad, there is no call to action. Inventorsgarage does not need image branding such as Nike, McDonalds or much of the F100. What it needs is targeted traffic, ie folks with ideas who want to run with them. Even better, folks with ideas, who are willing to buy services to bring their invention to market . A 125X125 logo graphic wont do diddly to bring the folks within that demographic over here. It likely will bring folks who are curious, and perhaps within that group their will be some… but why spend $ in order to get a mix of folks, when a redo of a logo graphic into a true ad might get a much more targeted demographic. So… gotta ad that to the todo list.

Along that line of thinking, if I step back and look at IG, and also look at my business page… I’m a plumber with leaking pipes at home. Neither were designed with commercial intent. IG is not really commercial, its more a personal/hobby site, albeit it does serve as a marketing funnel for business clients while maintaining the personal/hobby aspect. And my business site… well, it was created as offline folks kept asking for it. And while my focus is 80% hardware, the webside of things does get interwoven, and I’ve ended up doing a fair amount of website/commnuity consulting… thus it doesnt work out too well to tell folks they should have a website, when one doesnt have one of their own. (I hate to admit it, but i actually have a vbulletin license mostly just for client test and debug on my server at the office) LOL

The end result… the business site needs an overhaul. If a client said hey check this out, I’d go whoa… that has about a zero chance of converting. Too much of the content is more applicable to white papers, than actual calls to action. It also needs some serious work in the graphics and gui part (talk about zero whitespace, and really no clear calls to action, or clear whats in it for me to type of material). Ultimately it goes back to the beginning, it was put up in a weekend, as folks kept asking for it, and i needed to get it done. I should have put a plan together first as I always tell folks to do upfront… dont even think about coding or software selection, first figure out the overall vision, then from there make the more detailed decisions. Otherwise, its very likely that the tail ends up wagging the dog, ie the tools drive the websites function, rather than a vision for what the website should be. And yes, being a wiki driven site… it does lead to a lot of verbose subject matter. Of course the only time I’m not verbose is when we try to put a a whole website and a server on a 8 bit microprocessor with only 64K or 128K of ram.

Thus, more stuff on the web todo list… it will be interesting.

Gosh, it seems there are a ton of new wordpress themes, many incredibly cool to look at… at least initially. And then sure enough, testing across screen resolutions, browsers, and different functions they crash. Then add in, it seems more and more folks are encrypting parts of the themes. I really detest encryption, unless i know what the purpose is. The last thing i need is some type of embedded nasty in a theme. Thus, when I encounter such, I almost always decrypt, or at a minimum, fire up in a debugger and monitor whats going on. Usually, the encryption is to some type of back link generator. I was wondering why on earth someone would make a theme so poorly, and then put their name on it…. Talk about a disincentive to do business or link to them.

Well, I read something tonight, and now it makes sense….

Last week I suggested you create wordpress themes and release them for free with your link in the footer, the majority response was that not a lot of us can design and code wordpress themes, nor have the time to learn. A good alternative is to sponsor themes other people have created, another plus of this is that they will be promoting their theme, which essentially equals free promotion for you.

Apparently a lot of folks who are in a bit over there head are releasing wordpress themes! The thing is… graphics wise, many of those themes are top notch. Some incredibly beautiful layouts… just the underlying code is terrible. Considering how many times I’ve run across this in the past few months, I’m almost thinking I should add a wordpress theme fixer service to the web dev side of my business. Of course the problem is… just like the plumber with leaking pipes, I need to do a bunch of web cleanup of my own before I start trying to market said services to others. Might be time to put an edge on my own saw again.

As part of asthma management, record keeping plays a major role. Yet, compliance with such is often times not all that good in the short term, and for the long term, likely non-existant. Part of the issue is the need to manually write down readings from a peak flow device, the next part is reviewing the data either by oneself, or in conjunction with a medical dude.

The use of an internet enabled camera phone, ie Iphone or Android could drastically simplify record keeping, provide for some level of medical review, and thus increase compliance multifold, and potentially reduce the cost of medical services.

In a nutshell, custom code would provide the following:

• Reminder to take peak flow measurement
• Instructions as to take photo of peak flow device
• Image processing, likely offline, such that the numerical peak flow measurement can be extracted from the image data.
• Thresholds pre-determined via medical professionals based upon said numeric data as to course of action. Ranging from no-action to med changes, to head to the ER.
• Long term trend analysis, perhaps via an offshore version of Amazon’s mechanical turk (albeit comprised of medically trained personal)
• Periodic updates back to the phone screen as concerns asthma management.
• Provisions to do a datadump to healthvault, any number of medical databases, or even printouts, in order to share data with a medical care provider.
• Provisions to graph the data in any number of user customizations.

The above would hopefully provide the following benefits:

• improved patient compliance via simplified data taking
• improved patient care through better management
• reduced medical costs via offshoring lower level functions, reduction in the the need for onsite physician time, and reductions in un-necessary ER visits / hospitaliztion.
• much easier self reviews via graphical charts of the data, as contrasted with a worn tablet of numbers entered via a pencil.

Said code would not require new peak flow meters, nor a specialized cell phone other than a Iphone, or Anroid variant. The code for the most part only deals with the higher level phone functionality, thus providing for a range of potential phones, to say nothing of a reduction in obsolecence. The use of offshore medical reviewers offers significant cost savings, albeit there would be value in such a device, even if the iphone’s owner was just doing self reviews.

It should also be noted that any camera phone with the ability to email photos, and recieve SMS messages could be used is such a fashion, elbeit the computing power available is much lower, thus necessitating more offsite processing.

The first android phones started rolling out on the T-Mobile system not too long ago, perhaps more widely knowm as the google phone. Android is the operating system the T-Mobile G1 runs under, and its almost open source, thus why I refer to such as android phones.

Whats cool, is the ability to write your own apps, and the easy of entry for one skilled in software is pretty low. No hoop jumping to get in the google store, unlike the Iphone. And thats what makes this all pretty amazing… if one has an idea, its easy to throw it out there, and see what sticks. Even the average software guy with an idea can run with a google phone application pretty easily.

So here we go with the pedometer part. The andriod os provides for access to the phones GPS and motion sensors, plus it has the ability to send data. Now, whats cool is, one could use the motion sensor/gps combo to determine whether one was walking, running, on a bike or in a car etc. No more errant 50000 step days like occur with many pedometers if worn while driving on rough roads. Even cooler, talk about a great way to do caloric burn calculations including data on the type of activity. Even niftier, one could integrate google maps, and take the gmap pedometer to the next level and then some. Some type of social meetup functionality, for group runs and motivation could also be added, all under the android os, or even going further, a scheduler type program based upon some type of goal setting, perhaps even live coaching and feedback for runners. Then throw in todays economy, where few have any money to spend, adding optional pedometer functionality to a phone one already has is a win win for all. The Apple Iphone already has this in a sort of limited way… but the key are all the additional features made possible by a open design, as contrasted with Apple’s closed and much more tightly control model.

Revenue wise… processing group data, motivational stuff, coaching, and crowed meetups, would need to be done on a server no doubt, as running the GPS and motion sensors with a high duty cycle would be hard enough on the phones battery alone, apart from the fact, its a phone, not a dedicated computing device. In addition, server side processing lends itself to fairly easy monetization via advertising, or one could even add a subscription model in order to provide more detailed reports/coaching. Lastly, in order to reduce the potential of the andriod pedometer becoming an orphan like the old sport brain, include dedicated software as well. One could include a open source desktop application for user to capture and review their exercise data in a much simpler and lower tech way than the server based solution.

Brian has a way cool idea here, and its one with pretty low impact to business operations.. sort of a win win all around. Peak demand is indeed a killer, but even apart from peek demand, continuous demand at the wrong time of day can get pretty spendy… and it may well be its someone else on the grid running at peak, which impacts ones energy costs. The uility has to get energy from somewhere, and if they have to pay a fortune for it, even if its outside of ones owns peak demand window, costs will increase.

In addition, we know the grid as it stands is in pretty poor shape…. and yet opposition to grid improvements, to say nothing of the messy politics situation as concerns the grid, makes it unlikely that anything will happen in the near future. Or at least not until the grid collapses in on itself at a very bad time.

I also thought it was cool that sun used “Discovery Research” to publish Brian’s ideas, thus making it difficult for patent trolls to get in and ruin what could have a huge impact on worldwide energy demand. That is super cool.

In the last few years, we had laptop batteries, cell phone batteries, drill batteries, and now perhaps even the ipod….

Japan Warns Of iPod Fire Risks - Podcasting News.

There really is no rocket science to designing LiIon protection circuitry, nor manufacturing such… but the risk of botching it is significant, thus care is needed. And that care seems to be absent a great deal. Granted everyone wants to be cheaper than every other guy, and cost controls are key… but not when it comes to protection circuitry, esp where such could lead to personal injury or death if such fails.

Yet, repeatedly, we see LiIon cells catching on fire…. granted, its been a few years since I was a chief engineer, but FMECA (failure mode, effects, and criticality analysis) were a critical part of any type of product design where a failure could concievable lead to a fire and loss of life. Granted, some products dont need a full Mil-Std-1629A analysis, and others will find that standard a bit week… but some level of FMCEA is a must. Winging a design where the potential for fire or explosion exists is pretty egregious.

Of course, the product design itself is often times only part of the issue… errors in manufacturing, whether it be via raw material or incoming part errors, or assembly errors can seriously compromise even the best of designs. In todays world of lean manufacturing, 100% final test and validation is often times skipped, to say nothing of process verification, much less stringent adherence to a avl and incoming parts qual. If the stats prove things out, often times with all too low a sample size. the potential for, and the mitigation of such errors ends up going out the window in the interest of cost savings.

Granted, there does exist an standard IEC 62133 – titled, Secondary Cells and Batteries Containing Alkaline or Other Non-Acid Electrolytes – Safety requirements for Portable Sealed Secondary Cells, and For Batteries Made from Them, for use in Portable Application. But alas, it is dated…. IEEE, CTIA and UL also have standards in the works, or soon to be released.

I thought this was an interesting comment from IEC online news…

Li-Ion batteries can’t be sold without built-in overcharge protection and over-discharge protection. Refurbished batteries, or after-market batteries, however, can also cause failures, such as explosions and fires. These and other after-market batteries–imitation battery packs with false labeling—can be of lesser quality than those provided by original equipment manufacturers and may not be designed with appropriate safety circuits.

Hmmm… considering many of the fires seem to be coming from OEM products, as contrasted with aftermarket, obviously someone or something is asleep at the switch. Yep, I had an oem cell battery about burst into flame, apparently someone got a real good deal on counterfiet goods at the oem… which they caught, and then recalled, but that happened after my phone melted. Thank goodness i was around when it started to go.

I’ve designed in Li Ion batteries as well as chargers… its not rocket science, but it does require a concerted effort on the part of all parties to ensure something doesnt go wrong in a very bad way. I’ve even consulted on botched battery designs, where in something got done on the cheap… and they caught the problem in production when things started getting hot. That is no time to be playing redo, but better there than after a fire has occureed and someones live was lost.  A few days running a FMECA upfront, AND acting on the results of such is so much easier, than dealing with stopped production, a product liability lawsuit, or even the death of a customer or their family. This is nothing to mess with… its just too critical to wing it, or design and chuck over the wall. Holistic engineering, design,  and manufacturing is a critical part of this…. its not worth skipping some steps in the interest of a few bucks.

When Pitch Trim Systems Turn Ugly.

Indeed, they can get pretty nasty. I remember my boss years ago telling me, it took all his strength, plus those of a mechanic sitting next to him to keep the plane under control, when trim runaway occurred. Thankfully, they were able to get the plane stabilized, and then troubleshoot the problem and return to the ground safely. In a single pilot op, apart from the strength issue, not sure it would have even been possible.

That being said, pitch trim runaway has plagued airplanes for years, whether it be due to control failure, wire harness fatique, or relay lockups, and those were the simple ones…. As we evolve towards more and more fadac aircraft, and push greater and greater tech into the cockpit, concerns of such will likely increase, but also engineering is a lot more diligent and forward looking than the days of the past, where such systems were more electromechanical in nature.

Along that line of thought, is unintended acceleration in passenger cars. Certainly with the advent of cruise control, the potential for such has existed for some time. Although, granted, the old analog cruise controls of days long since past were pretty robust, and when they failed. the catastrophic mode was a statistical improbability. Most of the time, the system would just go dead, or worst case lock up…. not go intermittent as far as accelleration goes.

With todays engine idle control, emissions control, and even things like idle boost for purposes of power steering or ac acutation, the engine control computer, and throttle positioner servos, and play a much more critical role, as well as often function in more of a real time mode, and unlike kick up solenoids, are not mechanically limited to only small changes in rpm.

This does post a bit of a concern in two areas. First, from a EMC point of view, only a few models get tested, and it is hoped and assumed, that car A is a representative sample of a wide range of bodies using somewhat equivalent electronics and systems. Should there be an oddity, like the wiring harness mounts on the mid year 91 Ford Ranger which results in a huge susceptibility hold, that could come back to bite in a huge way. In 91, it was a no start issue, and thus safety was not impacted… however, what if an immuity hold pops up, and causes a fault in the throttle servo subsection. That could potentially be a life critical problem… and even more so, what happens as vehicles age…. harnesses droop, insulation degrades, and grounding potentials could end up all over the place, especially in the frequency domain.

Secondly is the issue of software, of which the vehile owner typically has little control, The only thing they really can do, is make sure their vehicle is taken in for scheduled maintenance, and especially so if they get a notice from the manufacturer. Almost all vehicles have flash memory, and thus it is possible for the car makers to update software long after a vehicle has left the manufacturing floor. Thus, while it may seem spendy to bring a car in for seemingly trivial maintenance such as oil, fluids, and a quick inspection… this is also where new firmware may be uploaded into the vehicles computer, and it a new release is available, and it catches some bugs, it could well be a life saver.

Thus, with EMC issues as a vehicle ages, as well as firmware issues, sooner or later it is likely more and more cases of unintended accelleration will occur. Fortunately, there is the brake solenoid interlock which prevents moving a gear selector unless the brake is on, and likely this will not only prevent user error as it has shown to do, but also serve as a backup safety to prevent unintended accelleration resulting in bodily injury or property damage. Two things are key…

1, the interlock should never be removed, sure its a spendy part, and prone to wear and early failure, but it indeed is a mission critical device.

2. When shifting from park to drive, ones foot should be firmly on the brake… not just toe pressure to actuate the interlock system. I caught myself doing that the other day, which is what led to a little digging, and resulted in this blog article.

Its not unlike the days when I used to really enjoy the yoke controlled electric trim… not only for inflight, but for pretty much all modes, that was until a buddy had a Dutchess freak out on him, and that brought to mind my bosses experience too. After that, I only used it in cruise, preferring instead to keep the system off below 1000 ft. Electronics in the cockpit and in the car are great convenience items, but they should never stand in the way to where safety is compromised.