This Martin Thompson talk is excellent. I've found time-stamping and handling GTD (good 'till date/time) orders consistently across nodes to be one of the most challenging bits of building an exchange infrastructure.
0.17 is a rule of thumb we use in diving. There are many like it which allow a little leeway in case there's a slip-up. pp02 can surely go higher than 1.2 (I know personally) before causing problems in some people. That said, I bet if you're climbing at 0.14 bar that you've stashed a pony bottle of 100% oxygen somewhere in your kit, just in case, no ?
It should be pointed out that the record depth in the area is not Tarek Omar's 203 meters as the article mentions, it's Nuno Gomes' 318 (https://en.wikipedia.org/wiki/Nuno_Gomes_(diver)). This, with all due respect to Omar who is indeed an incredible diver.
Also, the article speaks of mermaids so I'll just add on a lighter note that for a long while there was a toy mermaid tied to the over-hang mid way through the arch.
I don't recommend anyone "inject" trimix ;) But jokes aside : Trimix is an (artificial) gas blend that includes helium, which is an inert gas that your body doesn't metabolize (though their is still a lot of unanswered questions about its effects below 250 meters). Your body metabolizes gases (normally only Oxygen) according to their partial pressures in your body. The partial pressure of Oxygen at sea level is just 21% (there's an atmospheric pressure of 1 Bar and air is 21% Oxygen, 78% Nitrogen, and 1% Aragon and other weird stuff), but as you dive deeper into the water and the ambient pressure rises, so does the partial pressure of Oxygen and Nitrogen. Oxygen becomes toxic around (to be safe) 1.2 ppO2 and Nitrogen causes narcosis at a pp that varies greatly from diver to diver. As a rule of thumb, somewhere between 40 and 60 meters is where things get dangerous. So, to reduce the partial pressure of these gases, Helium is introduced, reducing the proportion of the metabolized gases in the total mix, and allowing the diver to go deeper before the critical thresholds are reached. To dive below a certain depth (about 60 meters), a mix is required where the proportion of Oxygen is less than the minimum 17% Oxygen at 1 Bar required for your body to function. We call these mixes "Hypoxic Trimix". A typical deep technical dive involves switching gases (different tanks) at least three times during the dive, and grabbing the wrong regulator is a deadly mistake. Technical diving is not for the distracted. It is, also, the slowest and most meditative rush you'll ever experience. Add overheads and caves and it really does feel like an excursion to an other planet. Speaking of space travel : more humans have been to space than below 200 meters.
You should, according to PADI norms, be offered a refresher dive. It costs $10 more and you spend 15 minutes with a dive master reviewing basic skills before a normal dive. The dive master also then knows to keep you close during the dive.
What instructors and dive-masters actually do (or should / are trained to do) for those dives is hold onto the first stage regulator at the top of the tank for the entire duration of the dive and never pass 12 meters. The diver never even needs to worry about buoyancy. We use to call them Lipton (tea) dives, because it was just dipping tourists into the water ;) They are absolutely 100% safe when done according to the requirements.
I'm an experienced diver (PADI dive-master, ACUC instructor, IANTD gas-blender and normoxic trimix diver, TDI hypoxic trimix diver and dive-master) with several thousands of dives of which many at 100m+ (330 feet) depths. I lived in Dahab where for 16 months I assisted training technical diving instructors in the blue hole and at other less touristy dive sites around there. The blue hole is in not a particularly dangerous dive site, but its popularity attracts the most cocksure types trying to prove something. The main attraction at the blue hole is the arch. It's a 40 meter archway who's apex is at 56 meters and leads from the cylindrical blue hole out into "the blue", where depths are insane and theres nothing but water anywhere you look. It's very beautiful, especially while the sun is rising, as it faces East. 56 meters is not particularly deep, but it's deep enough for any beginner and even most advanced divers to get seriously narced (drunk on nitrogen). It's also _just about_ deep enough for the partial pressure of oxygen to reach a critical point where oxygen becomes toxic. When it does, your muscles try to burn off the excess oxygen and you temporarily lose control of them, they shake, so your chin and cheeks start to twitch and you can easily drop your regulator. When the twitching is over, you involuntarily take a deep breath...
If you're an experienced diver, know your narcosis limits, know your oxygen toxicity limits, know your air consumption, havn't had a drink the night before, and are physically fit, maybe you can pull it off without trimix. Otherwise, it's an absolute beginner's dive with the right gas blend.
- management fees, margins, and available capital can easily be modelled properly
- you can easily set up constraints (like no fractional trading for your S&P example)
- you can set up a proper point-in-time database to avoid snooping (especially if you're using earnings reports or other fundamental data which is often actually released _after_ it's published release date...)
- you can set up regime-shifting simulation environments (various market conditions)
- you can avoid over-fitting if you're back-testing (with dozens of techniques, most notably : test once and forget about parameter optimization)
I would say that with paper trading and back-testing the serious problems are that :
- your orders don't show up on the book so no one sees and reacts to your limit orders
- your "filled" orders don't affect the book, so you're not affecting liquidity, so the market doesn't change in response to your trading
- your bot has no access to market micro-structure strategies and conditional orders (and if you want to trade fast or are placing big trades you need them)
These are the problems that make any simulation unrealistic, and they are fundamental. It's shadowboxing, which is not entirely devoid of value, but which is certainly insufficient on its own.
(I've worked as a quant developing strategies for several funds these past 15 years)