All posts by David

Social Media Click-farming — (New Republic Article)

Screen Shot 2015-04-21 at 8.09.22 AMA new article at the New Republic caught my eye this morning. It is about Facebook (and other social media) click-farming.

The Bot Bubble (New Republic)

I encourage everyone to read the entire article. Yes, it will require some time and attention, but I found it very informative.

For those without the time, or inclination to read it, here are some of my takeaways:

  • Facebook Like Farms are real business. The one in the article discussed is as noted, more a ‘start-up’ type business than the old spam-factories of days mostly gone by.
  • Paying Facebook to boost your reach can be dangerous and destructive to your marketing efforts. One case study explains how a music company basically destroyed their accounts usability by paying for increased traffic from Facebook itself.
  • Facebook plans to further dilute organic exposure of posts, specifically from business pages. Currently that organic exposure is approximately 6%; unless you pay Facebook for more. But that comes with it’s own pitfalls.
  • Marketing on Facebook is going to become more and more difficult for small business to effectively use Facebook to market and grew community. This gives me serious pause to a current initiative that is requiring significant development time in an attempt to integrate with Facebook.

April at Laguna Seca

April can mean unpredictable where here Northern California. Last weekend we had a deluge of rain. This weekend was a little different:

Saturday evening at Laguna Seca.
Saturday evening at Laguna Seca.

The work for this Sunday’s trip to Laguna Seca started in earnest on Thursday, with some upgrades to the Track Van.

First of was constructing a strong anchor system for the motorcycle wheel chock. Using a pair of E-Track 2×4 mounts, and a 36″ long section of 1.5″ angle iron, a secure, yet easily removable mounting system was created.
IMG_2044
IMG_2043

E-track mounting system
E-track mounting system

Another pair of E-Track anchors were added to the inventory, along with some more cargo straps for loading additional equipment (which came in handy as I’ll note later).
IMG_2041

Along with updates to the truck came upgrades to the bike; principally a new set of proper track tires on Friday. The new hides are Pirelli SuperCorsa in medium compound rear, and soft compound front:

Out with the old:
IMG_2032

In with the new:
IMG_2039

IMG_2040

Having done quite a few track days over the last year, I’ve come up with some items that I think are crucial to enjoying your day at the track. First of them is SHADE. Shade is very important here in California (I offer this to my friends in Seattle that I’m not sure have ever seen the sun). In the past I’ve been fortunate enough to have friends with nice easy-up awnings, but this outing was planned to be Solo, so I dropped $200 on a pretty heavy-duty setup at Costco. Although not as ‘easy-up’ as others (I have to remove the top each time to pack it) it’s big, and comes with side panels that were REALLY handy when the wind picks up at Laguna.

In the truck everything has a place, and everything is packed in it’s place; ready for the quick drive down the road to Laguna:
IMG_2051

The next morning, it was a short walk to registration and tech inspection. Not sure why the app was showing 44F, it was low 60’s (you might notice people are wearing t-shrts) at 7:45 AM.
IMG_2057

Once passed through tech, the generator was fired up so they would properly heat up the tires during the riders meeting, leaving the bike ready to go for the first session.
IMG_2058

With the generator and tire warmer setup on loan from CT Racing, it was a great day of riding!
LagunaSeca01

LagunaSeca02

LagunaSeca03

LagunaSeca04

LagunaSeca05

LagunaSeca06

By the last session, I was able to cut 6 seconds off my previous best lap times, clocking a 1:51.34 just before the bike ran out of gas entering turn 2.

A couple of the lessons I took away from this last outing:

  • A good set of tires are essential
  • Tire warmers are not hype, they not only keep your tires up to optimal temp they let them last longer and operating more consistently
  • A lap time timer is mandatory if you’re trying to develop good consistent laps… what I found was laps that *felt* fast were often slow, and the ones where I was relaxed, taking it easy, were my fastest by far. If you want to know if you’re really getting quicker, you’re lying to yourself unless you use a lap timer
  • Proper preparation for the day is important. I had a maintenance issue develop right before lunch and I had to scrounge up materials to build a brake bleeing setup, and get some tools loaned to me because the things I expected and brought tools for were not the ones that shut me down
  • Laguan Seca is windy at night

I’ll be doing this all again, May 10th!

Pinterest – Scaling the App (link)

A very good read on scalability and high-performance data management.

Over the last 10 years I’ve used all but Mongo and Redis to solve these very same issues, and had the same Findings.

A couple of the surprising lessons to me was how bad Cassandra was, and how good Solr is.

I hope you find this as interesting as I did:

Scaling Pinterest – From 0 to 10s of Billions of Page Views a Month in Two Years

Life and Death – confronting the reality of our existence

That sounds pretty ominous, and in a lot of respects, it is.

Yesterday was a strange day indeed. While working from home during the building emergency water shut-off, we noticed Police presence next door, and not the typical type.

Here is the reason: the Coroner had arrived to remove the body of one of the people living in the unit next door. It’s a little weird to know that someone had just died next door, just a few feet away:
IMG_1594_death

They are both aged, and have in-home nursing care. I guess when that is the case the Police have to very carefully investigate the death. There were 3 detectives, county Coroner technician and a patrol officer all going over the scene.

When you see detectives leaving with bags of marked evidence, you wonder if something foul had occurred.
IMG_1596_death

We later spoke with the complex maintenance man (for reasons related to the water shutoff), and he confirmed that they believed it to be a simple case of death by natural causes, but they have to impound medications, equipment etc. to make sure no crime was committed. He also confirmed that it was the first death in the building. Hm.. some consolation. :/

It was a very strange day, indeed. Godspeed to our neighbor. I hope they are now in a much better place, and able to get some deserved rest…. and no longer woken up in the middle of the night by the bar patrons across the street… :/

Track Sprinter: checking high-pressure pump

The number 1 suspected cause of the alternator failures, according to the Sprinter People I’ve queried, is a leaking high pressure fuel pump. So, even though I’m not identifying a clear indicator of a problem, it just seemed prudent to take the time to investigate it while everything else was apart.

There were a lot of steps not shown to get to this point. They all seem pretty straight forward, so I’ve skipped ahead to the part where I actually remove the pump.

There are three E10 torx head bolts that secure the pump to the engine block. Two are fairly easy to access and are located on the upper half of the pump assembly.

Here is one of the bolts being removed.
Sprinter_2015_01_20_05

The next bolt is located to the left, about mid-pump, very close to the connector you see removed in this photo.
Sprinter_2015_01_20_06

Here is a slightly better angle showing where the bolt is located.
Sprinter_2015_01_20_07

There are two electrical plugs that need to be removed. The one on the right slides out forward. The one on the left slides upwards. The left one is a bit of a pain to removed. I used some channel lock pliers to get an even pressure on the rear clips and gently wiggle it upwards until it came off. Take your time, breaking more parts while doing a repair is not a good plan.
Sprinter_2015_01_20_08

Plugs disconnected and ready to move onto the last steps. NOTE: The high pressure outlet has already been disconnected (top of pump). You can see the threads of the connector at top of photo. To removed these I needed an 18mm open wrench to hold the lower fitting in the pump, and a 14mm flare nut wrench to remove the coupling on the high pressure line feeding the injectors. That’s an important step. You won’t be able to remove the pump with them connected. What I didn’t remove when I was working on mine (maybe a bad choice) were the two lower pressure feed and return lines about mid-pump on the right (not seen here).
Sprinter_2015_01_20_10

Last of the three bolts is bottom center. As much as I’d like to do as little removal of parts as possible, I don’t really see a practical way to get at this bolt without removing the fan blades from the clutch. You just have to do it. I found the 3 bolts on the fan to be 5mm HEX socket (not torx). They were torqued pretty tight. They are a normal right-hand thread, nothing tricky there.
Sprinter_2015_01_20_11

And what did I find on the back of the pump? 2 of the 6 bolts had completely backed out. You can also see that there is diesel fuel on the back of the pump. This matches up with the failure mode described by others, and is the smoking gun (along with the turbo resonator) that I believe was killing my alternators.

All 6 bolts were thread-locked with blue and re-torqued, and pump re-installed.
Sprinter_2015_01_20_12

With the pump re-installed, next step is to re-install the alternator, turbo resonator, intake charge system and get a volt meter installed for diagnostics.

Progress is being made!

Sprinter T1N Failing Turbo Resonator

It’s a common problem, when the T1N Sprinters get a little long in the tooth. The turbo resonator (pressure side) can start to split, causing loss of boost and the dreaded Limp Home Mode.

Mine was not exempt from this plight. Below are some example photos of what the Turbo Resonator looks like when failing. This example shows the common separation of along the bonding lines. During high boost this can open and cause the loss of pressure, power and then a CEL fault which the trucks ECU detects as a dangerous condition for the engine and cuts severely reduces power.

This is what the part looks like. It is connected directly to the outlet side of the boost section of turbocharger. The rubber o-ring seals to the turbos (pressue side, not exhaust / impeller side).
dodge sprinter 3500 turbo resonator

Closeup of the gap formed at the bonding point. I’m told this is the typical failure point. In a lot of case you wont’ be able to see this unless the motor is under boost. In my case, it looks like it’s been failing for some time, it probably why the seller unloaded the truck $2500 under asking price.
dodge sprinter 3500 turbo resonator

Another area that I’m told is possibly suspect fails is the outlet side. I didn’t see any signs of a failure here, but it’s something to check if you suspect you have this issue.
dodge sprinter 3500 turbo resonator

Another thing to look for is leakage around the o-ring. In this case there is crud around the area, but the ring appears in tact and no signs of blow-by. It turns out the crud was from the primary cause of my reliability issues, from a totally separate part.
dodge sprinter 3500 turbo resonator

Fortunately, this part is not really difficult to remove. A 1/4″ wrench or an E8 female Torx socket (later bring the correct tool) and remove too small bolts when it attaches to the alternator bracket and it’s removed.

Also of good news is that a good, reported very reliable by the Sprinter community replacement, is available on Amazon.com for less than $30.

Track Truck: conversion back underway

While waiting for repair parts, and researching the possible causes for the alternator deaths… I brought over The Boy to help out with some cleanup and conversion prep.

The original flooring was actually in very good shape, other than the poor paint job applied by a previous owner. I set the boy to scrubbing it down with a high-tech degreaser and a lower tech broom. Some of the old loose paint easily came up but the rest seemed adequately adhered for the next step of the project; new flooring.
Sprinter_2015_01_18_01

The rubber flooring I planned to use came in rolls from China, so they had been in that state for a long time. To ease installation when that phase comes around they were unrolled flat and stacked under the plywood floor. Then I found some stuff to stack on top of it to make sure they were reasonably flat.
Sprinter_2015_01_18_02

Back to the headlights… as you can see, the still need a lot of work. This is the 2nd of the pair. Again, it will be treated with sanding and polishing.
Sprinter_2015_01_18_03

The other light is sufficiently refurbished for my needs, and was set aside during the previous day’s efforts. It doesn’t look too bad. But still not perfect. Note: I actually did one more round of this before re-installing them in the van.
Sprinter_2015_01_18_04

Back to the interior work. Getting the road noise levels down is an important part of this projects. If I plan to do long distance treking in this, I don’t want to be yelling all day, or fatigued from the noise. I did test it with an DB meter and the average level was 85db on a highway.

Here the poor-man’s Dynomat is being installed over the wheel wells. This asphalt roof repairing product is reported to have similar sound deadening properties as the vastly more expensive Dynomat. A small roll of this is available at the local hardware box for about $16.

Using a heat gun makes short work of complex curves. Easy-peasy. It’s just a little time consuming to cut the stuff, and press it into place.
Sprinter_2015_01_18_05

The first (and eventually failed) effort to attach the rubber to the wood. In this first round I used a Liquid Nails product. The specs looked good; very low VOCs; water cleanup… what’s not to love?

Placing pressure on the mats then maintaining with some stuff laying around the shop so that it bonds nicely to the floor. At least that was the plan I executed.
Sprinter_2015_01_18_06

Shifting gears a little bit, I removed the Turbo Resonator and examined it on my operating room quality work area. I have a separate write-up on this. In the end I determined that the part was failing and likely contributor to the alternator death syndrome.
Sprinter_2015_01_18_07

Back to the interior. With some help from The Boy we finished covering the wheel wells. The aluminum backed roofing seal didn’t fully cover the wells, so a less expensive product (Jiffy Seal) without the aluminum backing was acquired to complete the product. The Jiffy Seal was then covered with aluminum tape to maintain the same “look”.
Sprinter_2015_01_18_11

One done with the wheel wells, I popped back onto the internet to look at possible causes for the failure, the most ominous of which is a possible failure in the High Pressure injector pump (large black assembly on the right).

It does not look like a fun thing to remove and look for problems. I’m generally of the mind that if it’s not clearly broken or in need of refurbishing, I don’t want to mess with it. Some more investigation is underway.
Sprinter_2015_01_18_13

Below and to the left of the pump is the turbo charger. The intake hose as been removed and the high-pressure cold-side routing, which includes the turbo resonator has also been removed. Both the resonator and the pump are above the alternator and could possible cause it to fail should they leak fuel or high-pressure intake gasses.
Sprinter_2015_01_18_14

High pressure pump in close up. The black part is the front block, which bolts to a larger raw aluminum block with the drive assembly. It’s said that the pump can leak between the two plates if the bolts that hold it together, internally, begin to loosen. This is a known issue with the 2004-2006 model year.
Sprinter_2015_01_18_17

Next effort will be to consider removal of the high pressure pump. At least the front end is mostly stripped down and access will not be horribly difficult.

Track Truck: Diagnosing the illness

While doing some research on the problem and waiting for parts, I decided to pull out the headlamps to work on them.
Removed from the truck, they are easier to work on.

T1N Dodge Sprinter headlamps
T1N Dodge Sprinter headlamps

One diagnostic task, is to look for unusual current loads in the system. First stop was the headlamps, looking for evidence that some clown hat put in 100w bulbs or some other nonsense. What I found was just that sort of… nonsense. Luckily they didn’t over-watt the lamps, but at some point someone put in those inane blue headlamps (but they looks so cool.. no!). I didn’t detect this because the blue one was burned out. All the better. A pair of 55w H7 replacements will be procured.
T1N Dodge Sprinter headlamps

With the lamps out, I’m going to do some sanding and polishing on them to return some of the original clarity. Right now, they are much better after that first attempt, but they are not where I want them to be.

T1N Dodge Sprinter headlamps
T1N Dodge Sprinter headlamps

Some sanding with 800, 1500 and 2000 grit wet-sand, then a light polish with Mequires… they don’t look too bad. But I know they can be better.

T1N Dodge Sprinter headlamps
T1N Dodge Sprinter headlamps

To fully diagnose the electrical gremlins that are plaguing me, getting unrestricted access to all the grounds and the glow plug relay pack, goes a long way. So.. I started to take apart the front bodywork.
T1N Dodge Sprinter 3500

Neat Sprinter tidbit… this is the air intake (other side does not have this vent). This is a true cold-air intake. Air box is right there behind the headlamp (or where it would normally reside). Also, it’s possible to get at the grounds behind and under the headlamp with those parts out of the way.
T1N Dodge Sprinter 3500

After re-checking and cleaning all the ground table locations, and there are plenty, I focused on the glow plug relay. Typically the relay itself does not fair hideously, but what more commonly happens is a glow plug (or 5) will short out, causing a heavy drain (more than 20A) on the electrical system. 5 glow plugs shorted out is over 100A of current and would easily kill the 90A stock alternator.
Sprinter_2015_01_17_07

Based on information obtained from the various Sprinter forums, I fabricated (term used loosely) a jumper cable that can be used to test glow plugs for a shorted condition.
Sprinter_2015_01_17_08

Because the simplest place to test it from the glow plug relay harness, I unplued the 5 wire connector (1 for each GP).
Sprinter_2015_01_17_09

On one end of the wire is a spade connector that is screwed to the battery. I didn’t want to have to mess with holding both ends of the wire. On the other end, I crimped on a small finish nail. This is useful when testing the glow plugs from the glow plug harness.
Sprinter_2015_01_17_10

Poking the wire into each of these connectors here, I was able to determine that none of the 5 glow plugs are exhibiting the signs of a short. When I bought the vehicle the owner assured me the glow plugs had all be replaced in the last 2 years. It seems very likely that was a truthy statement. All good in this arena. But that also means I must continue to look for the cause of my woes.
Sprinter_2015_01_17_11

Since the bumper on this thing was in pretty bad cosmetic shape, I decided to remove it too for a little refurbishing at some point down the road.
Sprinter_2015_01_17_12

This little LED light really came in handy when working on things under the beast. Takes 3 C-cells and provides a pretty good amount of light, even if they LEDs are not the wide angle type. It was much easier than the conventional flashlights (due to built-in stand) and not hot and tangly like an old-school drop light (I really do have a heavy dislike for them).
Sprinter_2015_01_17_13

To keep the new battery charged up while I sorted out this mess, I re-purposed one of the 25′ charging cables from the MV Agusta, added the charging lugs to the Sprinter’s battery, and plugged it into this gem I picked up last month. Love having the computerized 1.25 AMP computer controlled charger.
Sprinter_2015_01_17_14

With the Glow Plugs eliminated as the likely cause of my troubles.. the next step is more drastic… checking the turbo system and fuel system components. Good times ahead.

UPI and Washington Post Twitter Accounts Hacked!

When I heard about a UPI report that the US Carrier George Washington has been attacked by a Chinese war ship, and was damaged, it smelled fishy so I started poking around the interwebs.

USA Today:

New York Post, UPI Twitter accounts hacked

UPI Responds

UPI website, Twitter account hacked

Press International’s website and Twitter account were hacked Friday afternoon, with someone attempting to publish false stories.

It started on Twitter, where six fake headlines were posted in about 10 minutes, starting about 1:20 p.m. Some of them were about the Federal Reserve; others contained a false report that the USS George Washington had been attacked.

Engaget

Pope declares WWWIII?
Twitter accounts for NY Post, UPI hacked to push fake WWIII alerts

Screen Shot 2015-01-16 at 1.01.34 PM

NAVY Times

I was actually alerted to this by the NAVY Times post that the carrier was not attacked and actually safely in port!
Screen Shot 2015-01-16 at 1.04.32 PM

Navy: China has not attacked U.S. aircraft carrier

The aircraft carrier George Washington has not been attacked, and World War III has not begun, despite what tweets from United Press International say, the Navy has confirmed.

The carrier is in port, not in the South China Sea, the Navy told Military Times on Friday.

Sprinter: Getting parts, and getting towed

Absolutely a day of ups and downs… so.. let’s start off where the day did.

Having just finished putting the truck back together again, it was time for a really good test drive. This time I picked Hwy 17. A 16 miles windy trek over the 1,814ft. Patchen Pass in the Santa Cruz Mountains (the epicenter of the 1989 World Series Earth Quake, mistakenly associated with San Francisco, 40+ miles to the North).

Hwy 17 displayed on Google Earth
Hwy 17 displayed on Google Earth

The old dog handled the mountain with ease. Passing slower vehicles on the steep climb was a non-event. Considering the type of vehicle, it handled the curves quite confidently.

Once in San Jose, first top was Club Auto Sport, where my trainer Isaac has his facility (Fit for Speed). Had a great workout, then headed out into the sun to start running some errands while in The Valley.
Sprinter_2015_01_1301

First stop was Harbor Freight to check out the finest in Chinese manufactured crap for American consumption.

The truck will need a set of wheel chocks, and these were on sale for $40 each, so I grabbed a pair. Don’t know much about them, but hopefully they are not complete crap.
Sprinter_2015_01_1303

Also needed was some rubber flooring for the back. The original plan (as you might have seen in the previous post) was to carpet all 14′ of the van. After some discussion with some other track junkies I know, a change of specification to rubber flooring was made. This stuff was not very expensive, so I picked up 4 rolls (enough to cover 6′ x 8′) of the stuff on the left ($38 total).
Sprinter_2015_01_1304

Next up are some ramps. These on the left are 8′ long and rated at 500 lb each. Nothing I own weighs more than that, and made in steel, they seemed stout enough for my purposes. A pair is $72, and that fits the budget nicely.
Sprinter_2015_01_1305

Ramps, rubber and sundries loaded up and ready for the next stop. However I spent so much time in Harbor Frieght, I needed to get rolling back home to complete some software projects. So.. heading to the dealer for a spare key and the RV shop to look for lighting, will have to wait.
Sprinter_2015_01_1306

Sprinter_2015_01_1307

And waiting I would be doing. A lot of it as a matter of fact. Not far out of Los Gatos I began to lose power, followed quickly by a batter light and a vivid reminder of my breakdown on Hwy 1. Recalling how suddenly, and with little warning it died last time, I found the first decent turnoff on the little mountain highway I could find, shut it down and… just felt pretty damn deflated.
Sprinter_2015_01_1308

Having just a few days before upgraded my insurance policy on the van to something called “Full Roadside Coverage” I ran up my insurance company and for the first time in my 50 years of life, tried out Road Side Assistance.

Very long day/story short.. they dispatched a flat bed heavy-duty tow rig from Santa Cruz, and said that the distance to back to town (about 22 miles) was ALL COVERED. I asked the operator if I could get it towed to my personal shop, and she said I’m covered for up to 26 miles, the distance to the nearest Mercedes Dealership that can service the rig. Excellent (considering I’m getting towed).
Sprinter_2015_01_1309

The tow truck driver was a nice enough guy. We talked cars and trucks all the way back through the hellish commute traffic (It was 3:00PM by the time we started rolling up the hill). As stated, they transported me all the way to my West Side garage, we unloaded and I put the van way for the next round of repairs.. :/
Sprinter_2015_01_1310

Unloaded all my stuff from the back….
Sprinter_2015_01_1311

Then unpacked the ramps to give check out the length and angle.
Sprinter_2015_01_1312

The deck height on the sprinter is 2′ 3″… which.. is pretty darn low for a vehicle capable of carrying 5000 lb. of cargo. I might have liked to have 10′ ramps, I’m not 100% sure I’m strong enough to reliably get the bike into the truck solo… (perhaps a winch is in my future?). But I’m going to go with these for now. The ramps also have holes in the sides and top that I could use to bring them together and make a nice solid platform for getting two bikes up into the truck, walking along side them.
Sprinter_2015_01_1313

There is a lot more experimentation to do, but for now.. I have to do some research and figure out WHY this things seems to be eating alternators like a kid loose in a candy shop after hours. Something just aint Right.