Archive for category education

Omega Seamaster PRO GMT (2535.80.00) Revisited (Update #3)

After the first two updates (both in this post), I got along fairly well with the SMP. It got put into my regular rotation and it was fairly accurate (~+2s/d). However, I noticed two things: (1) The GMT hand would sometimes get stuck at 8am. And (2) the GMT wheel did not click from hour to hour when I was quick-setting it.

Another minor issue was, that the crown seemed to be on the way out. It required a deft touch if I wanted to wind the watch, otherwise it would slip.The dealbreaker was the GMT hand.

I figured I would need to open it up again to sort the hand. I did not yet know what the problem was, but I thought that either the hand was loose or that the dial opening was uneven.

While I would have it open I would do some more maintenance – I bought a genuine crown a while ago, but it did not fit the tube in the rep (it has the screw on the inside, not out as is the case here). I also ordered a genuine tube, but I was worried that the opening on the case would not fit. The diameter could be wrong. I recently procured a push-fit crown removal tool, but I did not know whether the crown tube was push-fit or screw-in. Both of them exist in Seamasters. I therefore decided to test this on a backup SMP GMT rep case I have, to see how to do it, before inflicting it on a watch I am actually using. If i then borked the case I use, then I could always use the  backup case. I did not want to immediately use this one, for two reasons:

  1. Because the thread in the back is not sealing perfectly – it is water-resistant to 60m (with the new crown), but still, it bugs me that I cannot screw it tight.
  2. The case-back itself (more on that in a second).

I followed the excellent Archer’s guide on how to remove the crown and everything proceeded without a hitch. I used an aftermarket Clarks crown and tube (I know this is advertised as genuine, but it isn’t really. It is, though, half the price of the genuine crown and tube).

When I got the crown out it became apparent why the backup case was not watertight previously (the crown tube hole was malformed). It was also too big for the tube I had. I used marine epoxy, put a thin film around the crown and pressed it into place. Waited 24 hours, repressed the crystal, lubbed everything, sealed as far as I could and then tested it. It passed! But the only case-back that fits that case does not have a display window. If I borked up the case that I use currently, this one would work, however it would not be ideal. The 2535.80.00 has a display back (as indeed the case I currently use, does).

In the meantime, I also bought a genuine bezel with blue insert. Here is a comparison photo between bad rep, passable rep and genuine.

As you can see the replica and genuine inserts are subtly different, especially at the lume pip. The genuine insert has an additional steel circle around the pip. Note that omega does not sell the bezel and the bezel insert separately, so the ring needs to be replaced too. The replica bezel works on a spring wire, or with a little pip on a spring. The genuine has an insert with a raised piece of metal that works as a clicker. Below you see the three bezel backs. The middle one is genuine, the left and right ones are replicas.


I left this issue to one side for the moment, since I needed to completely disassemble the case anyway.

One suggestion put forward in the guide linked above is to put the case into boiling water in order to loosen the Loctite securing the crown tube in place. In order to do this, you need to remove all the gaskets (the one in the back and the one between the case and crystal). You also need to remove the bezel and insert, and the crown, obviously (because there is a rubber gasket in the crown itself). You also need to remove the bracelet, because the springbars have springs inside and the whole thing will get destroyed by the boiling water.

I proceeded with dissasembly. And removed everything from the midcase.

I boiled it for about 10 minutes, dried it off.

And then pushed the crown tube out without much difficulty. Note to self – the SMP cases, especially the GMT case, need to be put on the machine crystal side in, otherwise, they slip and break the pusher (I have now ordered several spare pushers:) ).


When I put the crown tube in, it did not fit perfectly, there was a bit of space between the case and the tube. Not much, perhaps 0.1mm, and it might be solved with loctite, but I would never be sure about its resistance to water ingress. Therefore, I mixed a bit more marine epoxy and pushed the crown into place. I now had to wait 24 hours for the epoxy to dry.

Therefore I could turn to the movement. I replaced the GMT wheel, so I got the click back. I replaced the datewheel for the hell of it and used a diamond file to smoothen the edges of the dial opening. I also put just a dab of loctite on the GMT hand once I had it perfectly in place.

I reassembled the movement and left it running outside the case, to let Loctite dry (with many of these cemicals and glues, I am worried that the fumes in  a watertight space (like closed watch) will stick around and slowly erode the hands, dial and movement). The other reason I left it outside of the case was that I wanted to see whether there was an issue with the hand clearances (a lot easier to reposition the hands while they are stil outside the case).

While I was at it, I also replaced the stem, because the existing one looks to have been previously damaged by the pliers. Not strictly necessary, but since I had everything out, I might as well do it.

While I was waiting for the tube to dry, I lubbed the gaskets and re-pressed the crystal. Note to self – How happy am I now that I previously bought the crystal gasket replacement(s)? I am very happy :).

I cleaned the crystal and re-visited the bezel issue.

Genuine used insert (before cleaning)

The genuine (After)


I do have the bezel assembly of the genuine case, but the rep case hole is only big enough for the spring wire, not for the little steel pole that might work here. Hmm… will need to sort it out. There are a few options:

  1. Drill a hole in the case and insert the steel pip (like Rolex has them). The hole dimensions are Diameter: 1.6mm, Depth: ~3.3mm.
  2. Buy a genuine disk spring. Does not apply for this particular midcase because it does not have a drilled path for the disc.
  3. Replace the insert only. While I was taking off the genuine bezel to drill a hole ( I had it on, but it was not flush with the crystal, so I removed it again), the bezel insert fell out of the genuine disc. So, 3 it is.

Here are the two solutions rep cases employ, on the left and right there is a hole and a spring wire (above the left watch case). The middle case has a larger hole and the pip and spring.

While I was at it, I realised that the genuine insert is just a bit too tight for the crystal. I needed to uniformly enlarge it, just a tad. So I did.

Once the case is ready and the tube to dry, then I’ll need to lube the bezel assembly and press it into place. Everything else is ready for reassembly as soon as the epoxy in the tube dries (I protect the hole in the tube with rodico/watchmakers putty).

A day passes. I re-visit the GMT.

The movement has cheerfully worked through the night. No need to do anything with it. I oiled the stem with (Moebius Synth 9415 lube).

The tube is firmly epoxied in place and is not movable at all.

I lubbed the back gasket and the crown gasket with silicon lube and tested the closed case for water-resistance. I did the proper test where I left the case under pressure for at least ten minutes and only then submerged it and released the valve.

The case passes without any issues. 60M WR. I would be happy to even go diving with it, but will for sure use it for snorkelling.

I set the bezel, glue the insert into place with a few dabs of gorilla glue (there are some adhesive strips available for bezel inserts, but they often don’t do a god job in my experience).

I reassemble the watch, and admire it.

Genuine parts:
– Dial
– Hands (except the GMT hand)
– Crown
– Crown tube
– Bezel insert

Rep parts
– crystal
– case and bezel ring (I have the gen ring, just wasn’t able to use it).
– movement and datewheel
– Bracelet and clasp

I add the steel bracelet, set the watch and put it on my wrist, where it is right now.



Omega Seamaster PRO 300 GMT

I am interested in watches. I like tinkering with them. I have a growing pile of parts that I hope to reuse at some distant time in the future. There is this fascination I have with Omega. At first I dismissed them out of hand. I thought they were ugly and pointless, especially the Seamasters. But then they grew on me. They are not particularly blingy (at least not the generation I am interested in) and they do not carry an insane price tag. Genuine Seamasters can be bought for about 2000 EUR, used, but not abused. That does not really break the bank in the world of horology. I have no idea how popular they are in replica form. I know that people are really interested in the bond versions (oh yes James Bond now wears an omega. I think he switched from Rolex in 1995 – Pierce Brosnan was the first Bond to wear exactly the watch we are discussing, in Golden eye). It is not that I am such a huge fan of Bond flicks, but I do like the Seamaster. It has to be blue, though. I have enough black watches.


Through time, I owned four Seamaster divers’ reps. They all broke at some time, or I destroyed them inadvertently. I started with the barebones, Seamaster PRO 300 with a blue wavy dial. It looked like this:

Omega Seamaster PRO 300 (SMP)

It was not horribly expensive and was reasonably like the genuine. The little cap (the Helium Evacuation Valve) on the upper left side is in the right position (above the 10th hour), the crown is long enough and the crown guards are ok. The dial is pretty crap, but you could buy a used genuine dial for not very much. The bezel is duller and the dot is not exactly right. The bezel ring is misaligned – the pointy bits should be aligned with the bezel triangle at 12 and with the crown. The red part on the seconds hand is too short, and for this model, the date wheel should actually be white. But I like it black, so there. Also, the crystal has one-sided anti-reflective coating and the genuine SMP does not have AR.

It was still a nice watch. It broke in three weeks. It would stop suddenly. And if I wound it, it would start again. The problem was not power retention. I changed movements, same shit. I later  realised it was the ring that holds the movement inside the case. It was too tight and it was bending the ebauchement. My initial reaction was to get rid of the ring, and put in a plastic one. That works if you don’t need to wind the watch. There is an ISA (1198) quartz movement that has the same hands dimensions as ETA 28xx / 29xx. So I bought it (it costs about $10) and put it in. IT was working just fine. At that time, I did not have the Water Resistance testing equipment, or the timegrapher. I assumed it was safe to swim with – I greased the seals and thought I was fine. I took it to the swimming pool and it was ok. I then went on a holiday on another continent and the first day wore it in a jacuzzi. The inside of the glass misted over, which means that some moisture came in. When I came back, I changed the quartz movement and sealed the HEV which was not sealed before.

After a year or so, I bought myself a WR tester. A that time I had two SMP’s, one with an ETA clone and one quartz. Both crowns broke. One would free spin unless held in a specific way (which was not too bad for the quartz and the other was eventually destroyed by me – I wanted to shorten the stem so I cut it to the correct length but when I screwed it back into the crown, it broke right where at the point where none of it was left. I read about it online and realised I could use a weak acid solution which would dissolve the stem (iron) but not the crown (steel). This takes about 12 days of constant immersion. What I didn’t realise was that the pusher into which the stem screws was also iron, so I dissolved that too :(. Wah wah. Then the other crown stopped working altogether. I first bought generic crowns that are shorter than the SMP ones. And they don’t have the spring needed to screw-in. They work on the quartz although it is clearly the wrong crown.

So, I figured, I’ll buy another SMP. And this time the GMT version, because I like GMT watches. I found a genuine Omega prototype GMT dial on eBay and won an auction. So, I had the dial. I then asked a dealer to source me a Seamaster GMT. They sent this photo for my approval:

There are issues with this too. The dial is crap (the 12 markers are misaligned, the shade of blue is wrong, and the logo has the so called happy feet) , but I had another, genuine, dial. The GMT hand is wrong, but I thought I could source another. The datewheel is white, which is actually correct, but I wanted to change it to black. The bezel ring is misaligned and the HeV seems to be sticking out a bit. But, generally, I could work with this. It has the correct crown, at least.

When I got the watch, I attempted to wind it and the whole dial started swimming. WTF. Something was rattling inside. Also, the crown did not screw in. DoH! I opened it up and one of the case washers was kicking around inside. The two that were supposed to hold the movement in, were still there. Three washers? it turns out that one was wedged between the ring and the case in order to keep it in place. Now, I had a problem. Regardless of what I would do, this one would not be Water Resistant, probably. I changed the movement and attempted to secure it. IT worked up to a point, but I figured I would rather shave the ring with a dremel and secure it with fasteners as it should be. I proceeded to just slightly distort the ring when I put it into a vise to shave bits of it off. Well ok, I thought, I’ll leave this for later and put in a regular movement, with another (non – GMT) genuine dial I had. I used this for a bit and it was working fine, however it was emphatically not water resistant (I have blown out crystals and broke one of them from the force of ejection). I felt this to be a total letdown – what good is a divers watch that will get destroyed if I took a shower?

So I had three cases, one working crown, two crystals, two destroyed case rings (oh yeah, I tried to modify another one), but I did have two genuine dials and six genuine seconds hands that I procured a year or two ago. I seriously contemplated buying a replica of the new ceramic SMP, which is very very good, but the new ceramic SMP does not have the wavy dial. And I think that that wave dial is actually pretty awesome. And I have a genuine one. Well, I could buy the ceramic one and swap the dial. I was about to do that, when a sale came up on a forum I visit often. It was for a high end SMP GMT rep and the price was reasonable. Because I got burned before, I contacted the seller and asked whether the crown screwed in (IT DID) and what kind of movement was in it (Asian ETA clone with the GMT module). I figured the movement will need replacing (but I have a mint one in my spare parts drawer) anyway. And the GMT hand looked right, so I could reuse that or buy another. I have the dial, already. I went and bought genuine hour and minute hands. The genuine seconds hand, I already have. I knew there would be some work to do, but I might actually end up with a water-resistant divers watch :).

The seller was in UK and delivered the watch in less than 48 hours. Very cool.

Here it is:


It has the correct display case-back.

I figured I would first put it onto a timegrapher, to see how bad it was. IF the amplitude was low and it was very slow or very fast, I would know that it was badly in need of a service. Boy, was I surprised.

Face up, +0 s/d, amplitude 284

crown right, +5 s/d, amplitude 274


Face down, +3 s/d, amplitude 273

crown left, -2 s/d, amplitude 261


WOW. This would pass official Swiss COSC certfication. It is almost as if it was not a clone. Wait. It is not a clone, almost certainly. It has the ETA stamp and the numbers would also support the claim that this is a genuine ETA movement. Cool. Excellent. I am not swapping it. The winding was slightly rough, I’ll clean and oil the winding bridge and the winding mechanism.

I need to work on two bigger things. The (1) the insides and (2) the case

Genuine seconds, minutes and hours hands + genuine dial next to the unsuspecting victim.


(1) The insides

out of the case.


I took the hands and dial off. I replaced the date wheel for the white text on black background. I so don’t care that this is not right for this model. I dislike the the white date wheels on dark dials.

Black DW already on.



Here is a comparison shot of the two dials:

Genuine on the right (on the movement), rep on the left.


The lume is uncomparable, the dots are different, even the GMT color is brighter on the gen. I then attempted to fit the hands and encountered a snag. The dial hole is not big enough for the rep GMT hand. And for some reason, I we are not allowed to buy the genuine hand. I mean this one looks identical to the genuine anyway. Well. Luckily I have a dremel and a dremel rig.



After several incremental drillings, the GMT hand finally fit. Then I attempted to fit the genuine hands and was totally unable to get the minute hand on. I suppose it is possible that I inadvertently bought the wrong set (but it is very unlikely as the hour would not fit either then. The eta dimensions for quartz are 120/70/20 for mechanical 150/90/25). Well, whatever. I put the rep minutes hand in and the genuine seconds. If I can be bothered at some time, I’ll fix that. For now, it’s fine.

The movement has hands on.

I left the movement outside of the case overnight, to see whether everything is fine and started focusing on the case.

I mixed marine epoxy to epoxy the HeV shut.




I left the epoxy to dry overnight and proceeded next morning.

Next morning, I took the clasp off and prepared the case for WR testing (greased the seals and gaskets). The movement was working fine, by the way, so it is ready to be put into the case.

First attempt










DAMN. The crsytal blows. But not the front. The back. The front is fine.


I add silicone grease, and try again. No luck, it blows again. I glue the crystal into place. Wait for an hour, and re-do. I know that I could have used the sapphire crystal glue (which I have), but gorilla glue dries more slowly and gives you some time to adjust everything before it sets.

Third attempt at WR.










Yep! Passed this time. 6ATM. I then spent about an hour removing the excess glue from the back crystal. I used pegwood.

I dried off the case, and put the movement back in. Now, only the last bit remains. Cleaning the winding mechanism and oiling the winding parts. I won’t do the whole service, because there is so much I could screw up and I don’t have the cleaning machinery, so I would need to clean by hand and while I’ve done it before, it is long and painful and the watch does not seem to need it yet.

Movement cased.

About to clean and oil


And the last few shots of the watch.


Addendum 09/10/2017

There are always things that bug you about reps. Some are solvable, some not so much. Once I’ve done everything I had to do, the watch worked fine for a few days and then I noticed that the GMT hand would get stuck every once in a while. I attributed that to the uneven hole in the dial and resolved that I would fix it. I had an additional pet peeve, because the seconds hand was a rep one, although I have four genuine ones stashed away in my watch drawer. I couldn’t use the genuine ones, because the seconds wheel pillar was too low, or the seconds hand tube was to short, take your pick. Since I would be opening the watch up, I would try to solve this, I figured. The rep seconds hand is too stocky, the back is too long and the red part in front is too short, compared to gen.

Once I opened the watch up, I took the hands off, and managed to break the tube of the rep seconds hand. Doh. So, now I needed another solution. I had a Planet ocean seconds hand, with long enough tube to fit, but that hand does not have a dot but an arrow at the end. It would work temporarily. I could not find any rep seconds hands for an Omega, and gen ones would be useless. I figured I could buy a rep silver Rolex submariner hand, with an extra long tube and then paint the end red. The back in the rolex has a dot and in the Omega it has a square, but still better than an arrow instead of the dot in front. I ordered the hand and the correct enamel paint.

Planet Ocean seconds hand

In the meantime, I enlarged the hole in the dial, but it seems that I borked the GMT hand that now just spun on the GMT wheel without traction (for obvious reasons – it got stuck on the dial, but the wheel was still turning, which meant that the GMT hand was enlarged just enough to lose traction on the wheel). OK, I thought. Since this is the old version of the GMT wheel, where it does not click by an hour every time you set it, posing difficulties if you are not setting the GMT hand on the hour, I figured I could try and replace this one for the GMT wheels I have in my spares drawer. This worked, because the movement is the same in both cases (modified ETA 2836). I replaced the four pieces that need replacing (the calendar wheel, the calendar cover, the day wheel and the GMT wheel – this has been extensively documented here, and here for example). And then I found something else out – the old GMT wheel has a 2 millimetre diameter, and the usual ETA dimensions are 1.8mm. So, this is why the dial did not fit initially. Also the GMT hand is now absolutely too big, I’ll need to replace it. Also, I now have a hole in the dial, that is too big. Not unsolvable, the hand covers it, but I can now push it too far in, which means the outer diameter would now get stuck in the dial opening. So a bit of tenderness is required. I don’t have any Seamaster PRO GMT hands. I have any number of other GMT hands but they all look distinctly different from the original. The closest one I have is the rose gold Omega Aqua Terra GMT hand. Which is roughly the right dimensions and correct shape, but is not completely red as the genuine. I fitted that and I like it enough that I’ll keep it for now. I could always repaint it if I wanted since I have ordered the red enamel paint in the first place (for the seconds hand). I had to also replace the hour and minute wheels, because the original GMT wheel was a different height (lower) than the ones I have, which meant that replacing the GMT wheel necessitated replacing the hour wheel, otherwise the hour hand had no purchase. Which necessitated changing the minute wheel because now the minute hand had no purchase (and the seconds pillar was deep in the movement anyway…). That also meant that I could not use the genuine minutes hand, as it has to much of a lip and would not stay on, or would, but would immediately fall off, when I attempted to press the seconds hand into position. So I put in a rep minutes hand (which has the wrong colour lume, but hey, best I could do).



I set everything up, aligned the hands, put in the Omega Planet Ocean seconds hand, and wore the watch like that for a while, but the solution kept nagging me (I would always know that the seconds hand is not correct even when I switched it for the Rolex one. I kept thinking about this and it occurred to me that it is possible to buy a higher seconds central wheel with pillar (ETA part number 2801). Apparently the one I had had H1 (height 1: 4.80 mm) and there are six more heights to go (all the way to H6: 6.05mm). That could work – I could use the genuine seconds hand! I ordered two H6 and two H5 seconds wheels and waited for arrival. I know how this is replaced, it is incredibly fiddly if you do not have a microscope – you need to catch four little spindles in the correct jewel openings on the brige and all of them are eminently breakable. If you screw in the bridge without proper alignment, you will break them off, only later realising you did this, because the movement would not work. I have screwed at least three escape wheels this way. However, the good thing is that you do not need to touch the calendar side at all, except for removing the seconds hand, of course). When the seconds wheels arrived, I first test replaced one of them in another movement, just to get my hand in and when this worked, I went to the Seamaster and manged to replace the wheel in question without screwing up the movement. I did need to re-seat the bridge a few times, but I did not bend or break anything. One note for budding watchsmiths – you absolutely need to remove the winding wheel on the barrel, because the seconds wheel goes under it! It is also smart to remove the whole winding bridge with the barrel, plus if you are up for it, the balance too. If you remove the escapement lever, it makes it even easier, but I have managed to screw this up a few times, with the lever never again seating correctly (lack of a microscope FTW), so I tend to avoid this. You do not need to do any of this (except removing the barrel winding wheel), but it makes it a lot easier to set the drive train. Also, while you are at it, you could oil everything up, and clean it too, if there is a need. You have everything out on the top side of the movement, anyway.

H1 on the left, H6 on the right (sorry for the bad closeup – it’s iPhone’s fault)

So, I replaced the H6 pillar for the H1 pillar without issue and put in the genuine seconds hand. Yay!

This is the whole process (I took photos of different movements I have lying around. I hope you don’t mind me not opening up the movement that is currently beating in the SMP GMT. They are all ETA 2836 derivatives. Before you start you need to unwind, or (if you don’t know how to unwind) let the power go out of the spring. If you don’t do that you will be breaking the teeth off some of the wheels and the rest will go flying around once you remove bridges.

Fully assembled movement. Auto-winder module needs to be removed. Two black screws there.


Auto-winder needs to be removed pretty much always. Note. When you wind your watch and hear a whirring sound as if your winder weight is freely spinning, that means that this bridge is broken. Once you have removed the auto-winder module, the bridges and the balance are exposed.

You need to remove the winding wheel! See how the seconds wheel is underneath it? If you remove the hand-wind wheel, be aware of two things: (a) it unscrews clockwise! and (b) there is a spring underneath which is prone to go flying at high speed. They cost 6 pounds a pop. Just saying. You will notice in the next picture that I removed it, but for this you do not need to. I removed it from a non-working movement, because I needed the spring that flew somewhere on a working one.


The bridge removed, but barrel still in place.


You could remove the balance wheel before or after you remove the drive train bridge. Or leave it in place. Up to you.


Once you’ve lifted up the seconds wheel, put the other in place and follow the same steps in reverse. Balance goes in last. Be careful to seat the the four spindles into their respective sockets. If you have removed the escapement lever (under the balance) then once you have seated the wheels and reattached the train bridge, they should all spin, if you move the seconds wheel. If you haven’t removed the escapement lever, then they should all move together. If there is resistance, you need to see which wheel isn’t sitting right. That is why it is smart, although not required, to remove the barrel at least, so you can check whether the wheels are seated correctly. If you don’t remove the barrel, then you need to reassemble everything, wind it up and if the balance doesn’t spin, unwind, and remove the winding and train bridges and re-seat. Repeat until it works.

While I was working on the movement, I removed excess oil (it was swimming in it. I do not know who works on these watches when they are preparing them, but they should seriously cut down on the amount of oiling. There seem to be only two extremes available in reps – (1) bone dry, no oil whatsoever or (2) over-oiling everything, to the point where wheels are literally covered with oil. Literally. Seriously dudes, get a grip. both (1) and (2) are bad for the movement long term – one results in excessive wear and the other in gumming up the works once it starts degrading. The correct amount of oil is invisible to the naked eye.

There were still two or three things bothering me.

  • The crown is slightly wonky. It winds, but it will go at some point in time. Omega rep crowns really suck. I sourced a genuine crown, but it does not fit the rep tube. I have yet to find a tube that fits (I’ve tried several). Also, on most Omega reps, the crown tube is pushed into the case, but on this one, it seems to be screwed in. I bought the genuine tube for this model, which is also screw in, but it does not seem to work with the genuine crown I have. I know, this is all confusing. Simply put, the crown will go at some point and then, I’ll have a problem :). It won’t be unsolvable – I have another SMP GMT case available, that has a crown that does not screw in, which makes it not watertight at all. However, I have recently acquired a crown press and I have a rep crown that screws onto the tube, but does not screw into the case. This might work as a replacement case if the one I have fails. I would need to swap the bezel insert (because the current one is a lot more accurate than the other one) and I would have to see whether the display case-back fits (on the genuine and the case I am using there is a display back) on the other case, there isn’t. I also have another SMP case, but there is one distinct difference between the GMT and non-GMT cases – the crown guards.
  • The GMT hand is not correct, but I kind of like it, anyway. So, I might let this stay as it is, or repaint it at some point. For now, I am fine.
  • I am worried about waterproofing – the dodgy crown, etc. So, I took the movement out and retested it, this time properly – left the case under 6 atm of pressure for at least ten minutes, before lowering it into the water. It passed! So now, at least I am reasonably sure that there won’t be an issue there (I started marking the insides of the case when and what pressure they pass the WR test. Otherwise I keep forgetting which watches I can swim with, and which ones, I cannot).

Well, everything back to normal. The movement oiled, the hands seated correctly and waterproofing done. The date flips between 00:00 and 00:01. Here are two more  wrist shots:








About academic fraud

Image downloaded from Keble College Oxford’s web-page

I have done a presentation for a web-security firm in 2011. The presentation is here, and it is live, my voice and picture included. I talked to a crowded room. I was nervous but it went fine.

Recently, I needed to find a reference about Steve McQueen using the word scam for the first time. There was an interview, it was published in Time Magazine in 1963, but I couldn’t recall who the author of the article was. So I googled Steve McQueen and scam and Time Magazine. There was a hit and the wording seemed familiar. So I clicked on this link. If the link is removed, as I suppose it will be, after this becomes public, here is a screenshot of the page and here is the locally stored pdf (fun starts at third slide).

Interesting, right? Amon even took the pictures from my presentation, not only the text. I would be interested to know what the Royal Holloway has to say. I shall contact them later and post their reply here. If we ignore the first two slides, the only thing Amon did, was replace my name with his and changed the background. Not bad, right? Certainly much easier than going to the trouble of researching this.

Let’s look into Amon Sanniez, the plagiarist. Here is his linked-in page. It seems that years of expertise in security and in ownership (heh) have landed him a job a company that does pest control. He used to work for security firms and has lots of qualifications, but not much academic clout (a BSc was the end of the road). He does list my presentation as his publication on linked-in. Here is his blog. The subset of the things he writes about is in my field of expertise and he is not completely on the ball there. The things he writes about there are presented with a simplistic point of view and they are often bounded by a lack of knowledge of the field.

The biggest irony is that Amon replicated my presentation but I now know where I was mistaken and he does not. Several of the things I say are just plainly not true now, and were problematic then, but we (the real scientific community) didn’t know it at the time.

Oh well, best of luck with your future prospects Amon, I am not too bothered about you stealing my work, because I moved forward since, and you, well, not so much.

Note: this is what I sent to the Royal Holloway:


Name is dr. David Modic, I am a research associate at the Security Group, Computer Laboratory, University of Cambridge. I apologize for simply contacting you out of the blue, but I think you should be aware of the following:

In June 2011 I gave a talk at BAFTA for the company Websense on the topic of Psychology of Internet Fraud. Here is a link to my talk: (with my photo and audio narration).

Two days ago I was looking for a specific reference and I googled it. I found the reference in a presentation by Amon Sanniez, here:–The-Psychology-of-Scams-and-Malware.pdf

When I looked a bit closer, I realized that this was my reference and the whole thing to be a direct copy of my presentation with only my name replaced and the template background changed. The first two slides are not mine, but all the others are, word for word, including the clipart. Feel free to check both links above.

 I looked at the event and it turns out it was held by the Royal Holloway Security Group in September 2011, if I understand correctly. The slides are hosted on the Royal Holloway pages. Your group was of course not to know that Amon Sanniez plagiarised my research and in the grand scheme of things it does not matter much anyway, but I thought you should be aware of this. If you paid anything to the presenters you were also scammed, in a sense. I guess there is irony in talking about fraud by committing academic fraud. I also realize that mr. Sanniez has no affiliation with RHUL.

Anyway, I just thought you should know. I don’t have any expectations about what happens next, it is completely up to you if you want to pursue it in any way. Please note that if I found plagiarized work hosted by Royal Holloway in about 30s  through google, others might too, and that may reflect badly on the RHUL, even though I don’t think any of this is in any way the security group’s or the University’s fault.

I am sorry to be the bearer of bad news.

Kindest regards,

— David


Improving the science

Here is a post about how science could be improved. To sum up:

(c) The New Yorker

  • Stop people from hitching themselves up on work of other, junior, researchers (stop ghost or honorific authorship).
  • Rethink REF and implement other measures of academic merit
  • Support replication and move away from positivistic science.

I like this. First of all, I agree that things need to change.

I am worried that it will never happen, though. Or, if it does it may be at the expense of junior research staff – Let’s say that someone is on a grant. The Principal Investigator (PI) for the grant got the money in and is employing them. If the researcher publishes their research and cites the PI as a co-author, what does that mean? Is it an honorific authorship or a ghost one? Even if the PI did nothing much to contribute to the article, should they be removed from the paper? If they did not secure the grant, there would be no salary for the researcher and the research would not happen. So, the PI is instrumental to the research in a very concrete way, even if the data collection, experimental design and write-up was not primarily (or not at all in some cases) done by the PI. I am not talking about myself, to be clear. I have absolutely nothing to complain about in this respect. 

Furthermore, if PI’s are removed from the publications that they enabled but did not invest much work in, that would mean that they would have to do much more research by themselves and not have much time to spend on preparing grants and thus there would be less opportunities for junior researchers to be employed. In the climate where it is hard to get a postdoc as it is, this is nothing to wish for. 

We know how these things work. As a junior researcher the options of securing grants on your own are fairly slim to non-existent as it is now. So, taking away the incentive to be employed by more senior researchers is not a brilliant move. Ah, I hear you say, you only need to be excellent and it all works out. Right? Well (a) at my place of work, everybody is excellent, otherwise we would not be here. So beating the internal selection process is a bitch and your work that receives worldwide recognition would not necessarily secure your future here. And (b) What good is your excellence if the only job you can hold is to be a nightwatchman at Tesco, because there are no grants available? That job does not offer many opportunities for you to showcase your research skills. Or leave you much time to practice them.

In my opinion, the notion that doing research for research sake and not in order to keep holding a job or increasing your viability for grant success is very nice and something to strive for. In the current climate, I think that it is somewhat romantic, though. How do you get the grant giving bodies to stop assessing the grant proposals on the strength of publication counts and Impact Factor (IF)? What would a replacement metric be? How often you appear in the newspapers? We already know from some Dutch research that personal fame is a stronger indicator of getting funds than publication count. That is even worse than trying to estimate merit, even if the current measure is flawed. It just makes you more likely to media whore and also introduces large inequality across applicants. I have been solicited by the BBC a number of times to do stuff with them. And we do collaborate every once in a while. How likely do you think it is that I would have been contacted by the Beeb if I wasn’t working for Cambridge? So, those of us who are in some ways in a privileged position, would really have a leg up on those who do not get those opportunities based on location or (lack of) prestige of the institution.

So, what should we do? Things are broken now, but they function to a certain extent. From a purely economic perspective, we should think about incentives offered to Universities and their senior research staff. Only people who do not work in academia in 1st world countries make the mistake of thinking that University Teaching Officers (UTO) are doing their jobs out of love. Most of us like and enjoy our jobs, make no mistake, but the discrepancies between the industry and academia are already huge. My starting salary in the industry would be about 2-3 times higher than it is in academia. The job security is about the same in both places at least until you are an UTO. After, it is slightly harder to get fired from a University, but you can still manage it if you mess things up sufficiently. So telling senior researchers that their jobs will be less secure (as a consequence of lower performance ratings, which is a consequence of lesser ability to secure grants, which is a consequence of their names being dropped from the grant publications), but that is OK, because they should simply love science, would not be unopposed, in my opinion. Saying that something should be done and proposing these measures without wrapping them in a bullshit sandwich is not a good way to go. If something makes logical sense, it does not mean that it will be accepted. Perhaps this is what scientists should actually learn.

About educational policy

American Education is in the Dumpster

American Education is in the Dumpster (Photo credit: brewbooks)

There is an article in the Guardian, titled: “Advent of Google means we must rethink our approach to education”

I generally agree with the premise that we could change the education process, but I think this article is somewhat skewed.

1) There is a difference between replication and synthesis of knowledge. It might be OK for the vast majority of the population to be unable to think for themselves and just rely on others to do their thinking for them, but – for example Stephen Hawking cannot google his findings, before he discovers them. In order to push experimental physics forward, he needs to to understand the theory behind it and in order to do that, he needs to be familiar with it and perhaps not on the level of Wikipedia (my additional question is – what if Wikipedia packs it in after an unsuccessful funding campaign? What happens with all functionally illiterate people who have not written a wiki page in their life, and, honestly, also couldn’t write it to save their life, if all they can do is to replicate what is already there). My point here being that it is just fine if a person is literate and can multiply 6 by 7 without using either pen and paper or a mobile phone. And, oh horror, you need to learn this by heart. How the hell would making people completely and uncritically reliant on technology make them more advanced? The suggestions here seem to me to lead to the absurd position where, in order to read a webpage, you would need to consult somebody who could actually read (for example, the village seer). And how is this progress? And what happens when the last of the ancient folks who still learned how to read, dies?

To keep Stephen Hawking out of this mess – what about a more personal example? I use a bibliographic manager for referencing. I don’t penalize my students if they use it in their essays, but I certainly do not encourage it and I don’t tell them that software like that exists (and they mostly don’t know it does). Why? It is not because I am a sadist pig bent on making these poor kids suffer. It is because I know how to cite without help from software. As a consequence of this ability, I can tell when my bibliographic manager gets something wrong (sometimes it does) and I can fix the mistake, because I know how to do it correctly, by hand. And my students don’t. Yet.  If they start by using crutches from the get go they will never be able to rely on their knowledge and ability. What happens if the referencing manager they use stops being actively developed and disappears from the market? What if it changes significantly? The students ability to produce academic writing will be severely impaired. My won’t, though.

You see this is not even the difference between being able to adequately play Bach’s Brandeburg concerto nr. 1,  and composing the said concerto. It is actually the difference between learning to play every note by listening to the composition and watching youtube videos on how to do it; and being able to read musical notation. I am not even at the creative part yet, I am saying that getting people to read digests and summaries of stuff and making them believe that this is all there is to human knowledge, is sad and misguided.

I am talking, therefore, about blind faith in technology and about reliance on somebody else’s ability to think. I am talking about Google search algorithms that are not necessarily constructed in a way which would give you optimal results, but are more probably constructed in a way that maximizes profits for Google. Lots has been written about that. But, hey, people educated in a way proposed by Sugata Mitra will never know about it, because they will be learning about freaking mobile phones, when they are nine years old. I realize I am simplifying a bit.

2) Mitra writes about how the teacher encourages kids to learn through situation and through interaction in-group and over the Internet. Well. My question is – does the teacher study the subject they teach in a classical manner or do they also simply Google it during class? Ok, how does this work? In the Guardian Column the subject is signals in mobile telephony and the teacher starts by having students look at electromagnetism. How can you make sure that the teacher will start with electromagnetism? Does the teacher even know about the connection? Where from? Isn’t this a case of blind leading the blind? If the teacher actually has a semblance of knowledge about the subject they teach, are we then saying that some people should be educated and some don’t have to be? Are we saying that it is suddenly not a problem that some people will be left uneducated? And is the idea that I as a parent should decide that my kid will be left to do menial jobs, where the ability to reproduce, but not actually create will be the only thing required? I think testing the validty of  this idea would be quite simple – I propose two observational experiments:

a) Let us see whether Sugata Mitra is content to let his kids not have a classical education but are instead taught how to google and how to wait for others to come up with ideas (i.e. work in a team, where individual performance is not assessed and where the accolade is shared regardless of input).

b) Give the parents a choice where they can either send their kid to school where they are actually taught something or a school where they twiddle their thumbs and sorta vaguely learn about how mobile phones work. And let them decide.

3) What the Mitra does not seem to get is that teamwork is effective when people in the team are capable of independent thought, not when they all wait for somebody else to provide an idea. Clueless people in a group are a herd, not a synergy. That is not progress, it is a regression. If you skip the step of enabling people to be creative on their own, you will only get drones. And I am not content to do this to my kid.

4) There is a nice slant to this – These kinds of programs are run out of UK into India and Africa. The underlying current is that the kids there are not expected to attend top level UK Universities or prep schools in the first place. Thus, you see, these kids are ripe to be experimented on. This will not be foisted on British kids, oh no, but these poor savages deserve their chance, right? They will never be as good as us, but they are good for a few papers and for the feeling that we are doing something  good for them. Oh wait, there could also be a grant in this. Right?

Enhanced by Zemanta