Express Failure In Logarithms: Fill & Download for Free

Absolutely.Don't misunderstand, I'd be the first to laud the importance and accomplishments of math. However, as a recent graduate who has seen numerous CS theses & research papers I can attest: it's certainly possible to do research and even get a PhD for that without possessing love of mathematics, just working knowledge of its applications.Just keep away from theoretical computer science. Theory CS is just a form of math; don't do that if you don't enjoy it. Do research in systems computer science instead. (There's also a big gray area of in-between that contains specializations likes algorithm design. The level of mathematical sophistication required for everyday work in these disciplines varies greatly, and depends on the particular research group you haven happened to get yourself into.)I have seen (and read) numerous theses that did not use math beyond basic probability theory and statistics (for statistical significance testing). They were on practical topics such as:programming abstractions (such as: for whole-network or network-region programming; for concurrent programming; etc.)programming language and programming toolkit designnetwork protocol designembedded operating system designthe "human side" of software engineering (collaborations, risk management, best practices of software engineering etc.)program testing"architecture" of somethingfield deployments of research-level software and the lessons learned from it.Some of these theses were award-winning, both at national and European levels. Good (and respectable) research is about finding and validating good research ideas, not some kind of contest between math nerds.What if you suddenly find yourself facing a problem that cannot be solved analytically without graduate level math skills? I personally have used all of these tricks:1) Research the literature. If it's a real and interesting problem, chances are that some mathematician has solved it already.2) Find more mathematically-inclined colleagues and ask for their advice. Chances are they will be happy to help!3) Don't solve it analytically! Use computer simulations or numerical methods instead. Approximate solutions in general are accepted in system's CS.E. O. Wilson, one of the founders of modern evolutionary biology (not CS, but math heavy-subject still) has a refreshing perspective on this:Mathematics [is] the great bugbear for many would-be scientists. I mention this not to nag but to encourage and help. I mean in this letter to put you at ease. If you’re already well prepared—let us say you’ve picked up calculus and analytic geometry—if you like to solve puzzles, and if you think logarithms are a neat way to express variables across orders of magnitude, then good for you; your capability is a comfort to me. I won’t worry so much about you, at least not right away. But keep in mind that a strong mathematical background does not—I repeat, does not—guarantee success in science. I will return to this caveat later, so please stay focused. Actually, I have a lot more to say to math lovers in particular.If, on the other hand, you are a bit short in mathematical training, even very short, relax. You are far from alone in the community of scientists, and here is a professional secret to encourage you: many of the most successful scientists in the world today are mathematically no more than semiliterate. A metaphor will clarify the paradox in this statement. Where elite mathematicians often serve as architects of theory in the expanding realm of science, the remaining large majority of basic and applied scientists map the terrain, scout the frontier, cut the pathways, and raise the first buildings along the way. They define the problems that mathematicians, on occasion, may help solve. They think primarily in images and facts, and only marginally in mathematics.You may think me foolhardy, but it’s been my habit to brush aside the fear of mathematics when talking to candidate scientists. During my decades of teaching biology at Harvard, I watched sadly as bright undergraduates turned away from the possibility of a scientific career, or even from nonrequired courses in the sciences, because they were afraid of failure in the math that might be required. Why should I care? Because such math-phobes deprive science of an immeasurable amount of sorely needed talent and deprive the many scientific disciplines of some of their most creative young people. This is a hemorrhage of brainpower we need to stanch.And especially this:It is far easier for scientists to acquire needed collaboration from mathematicians and statisticians than it is for mathematicians and statisticians to find scientists able to make use of their equations.

To date of 4,783 Boeing 737MAX on order, 377 have been delivered. That means odds are that 1 in every 177 MAX flying will crash. This aircraft avoided the FAA certification process through a legal loophole called Grandfather Rights.Even though there is not a single mechanical system left on the 737MAX that has anything identical to the original Boeing 737–100 series first launched in 1967, it was allowed to be flown on the certificate granted 50 years earlier.Design work on the MAX began in 2011. The advantages of this MAX design over the previous B737–800NG centred on a LEAPS-1B engine. This engine had a bigger Low Pressure fan with blades which could change angle of incidence thanks to a novel gearbox design. This feature gave these engine 15% superior fuel efficiency over a contemporary equivalent.Rival A320 neoIn 2006 Airbus started a project to re-engine their A320 design with a new much more fuel efficient CFM International LEAPS engine. The key feature of this engine was a gearbox in the Low Pressure fan which changed incidence of the fan blades much like changing gear in a car. This was how these engines could operate more efficiently. Airbus saw the opportunity from these engines long before Boeing did.Boeing’s rival aircraft the A320/A321 neo use LEAP-1A engines or in some variants the Pratt & Whitney PW1127G employing a similar geared fan concept.Boeing realized their 737 model could never compete with economics of the A320 neo, so back in 2011 a decision was made to rush into production a version of the popular B737NG with LEAP engines. Many corners were cut in terms of engineering solutions and safety certification.The biggest problem for MAX was only discovered during test flights was longitudinal instability. The MAX breached FAR part 25.173 regulations on stability.FAA certificationFAA who had been underfunded for many years could not manage to keep up with certifying all the aspects required so delegated safety certification to manufacturers. Certification is essentially testing an aircraft and all its systems to make sure they all work safely as intended.Boeing deliberately exploited a legal loophole with FAA that by not changing the name of the aircraft from 737, the MAX could be launched as a mere variant on an assumption that the variant would only have slight modification from previous variants.The problem with this however is the first version of 737 -100/200 models had vastly different engine design, smaller and more primitive airframe. The 737 then went through a significant change with larger engines to become the 737 Classic -300/-400/-500 series. This is the real ancestor of the MAX and already the design was much different to the first certified version.Enter MAXThen came the Next Generation or 737NG family -700/-800/-900 series. The MAX is in effect modified from the NG with more sophisticated avionics and the much larger LEAP-1B engine. The addition of this engine made the aircraft unstable. The engine was heavier and further forward affecting balance on landing.In the take off and climb this engine posed the opposite problem. It tended to push the nose up risking a stall in the climb. Boeing only realized this very late when they began prototype test flights. For that reason MCAS was a late addition to digitally manipulate the aircraft’s flight characteristics.Boeing had to modify the Boeing 737–800 to accept a larger engine, yet remain within constraints of the undercarriage & wing ground clearance. Undercarriage still had to attach the wing at the same structural points. The legs still had to retract into the same wheel housing. These features couldn’t be changed without a major redesign.Boeing’s automatic trim for the 737 MAX was not disclosed to the Pilots - Leeham News and AnalysisBoeing performed this feat by pushing the engine and thus mass weight further forward. Pushing the engine’s weight further forward causes a gyroscopic effect around the pitch axis in steep turns & steep climbs which change aerodynamics Now because it is nose heavy it does not naturally want to stall or flare upon landing. It has different stall characteristics. This was only discovered by test pilots flying it for the first time.Boeing’s New Jet Hit Problem in Tests Before Fatal CrashBoeing elected to fix the problem with MCAS which electronically manipulates flying controls to change the pitch trim and mimic the previous control feel of the 737NG aircraft. Boeing’s selling point for the MAX was that crew familiar with the NG series would not need expensive retraining to convert to the MAX.AOA sensor blamedProblem come however when MCAS responds to incorrect or conflicting flight data as happened on JT610. It is still not conclusively understood what went wrong on Lion Air and so far all we have is a working hypothesis blaming the Angle of Attack sensor.What is not in dispute is that JT610’s flight computer received conflicting Angle of Attack data. What is disputed is that faulty AOA sensors caused this faulty data. It may be that the data was corrupted when digitally encoded rather than the sensor itself being faulty. This poses a danger for re-certification if Boeing & FAA make false assumptions how to cure this?Boeing say LionAir JT610 was caused by a faulty Angle of Attack sensor (AOA), however I dispute this. My contention is that this was a digital signal encoding failure by the Digital Flight Data Acquisition Unit (DFDAU) possibly related to sensor logarithms used to trigger MCAS.I say likely this caused MCAS to activate features out of context and fight against pilot actions. Boeing call it a simple software problem. I say Boeing is wrong and over simplify the issue.Boeing 737 Max Major Design Flaws, Not a Software Failure - Mish TalkBoeing 737 Max approval documents subpoenaed by fraud unitSeveral faults experienced by JT610 or previous flights could not be directly influenced by AOA sensors. We know that following wild rides as JT2749 and JT775 the same aircraft had a new replacement AOA sensor fitted.JT775 was the last flight performed with the original factory fitted AOA sensor. In theory after this AOA sensor was replaced with a brand new sensor the problems should have stopped, but in fact they continued.Control Problems experienced after AOA sensor replacement:It next flew again as JT776 after the AOA sensor was replaced, but still suffered flight control problems. Then as JT43 from Bali the problem struck yet again. This time on JT43 the presence of a pilot from a sister airline, Batik Air in the cockpit saved their flight.Pilot Who Hitched a Ride Saved Lion Air 737 Day Before Deadly CrashOn JT43 an extra pilot riding in the jump seat is said to have intervened and disabled MCAS saving that flight. Why Lion Air allowed other pilots to keep flying the aircraft without briefing all pilots is a question for Lion Air management now?The Jump-Seat Pilot Who Saved a 737 MaxA day before it crashed, Lion Air's Boeing plane was reportedly saved by an off-duty pilotFinally the doomed MAX airliner operated JT610 flying with a brand new AOA sensor fitted after JT775, but still the wild rides and control problems continued.Thus I say it was not the AOA sensor itself which caused these crashesWhere this went dreadfully wrong on Lionair 610 was MCAS created a cascade of failures with Runaway Stabilizer Trim. Pilots are aware how to disconnect Stabilizer Trim, but a control yoke override feature present on the 737NG which allowed pilots to cancel a Runaway Stab Trim with a firm pull on the control yoke was deleted in the MAX design.What baffled pilots on Lion Air and which was different from anything explained flying previous models or expressed in their QRH handbook was that the Stabilizer trim would resume pushing the nose down every 10 seconds or so.Why this became dangerous is because of of a situation called elevator Blow Back in which eventually air pressure overcomes the strength of actuators to push the elevator back to level. At this point pilots on JT610 and presumably ET302 as well each lost control and their aircraft entered nose dives.None of the modifications proposed by Boeing for re-certification, can overcome Elevator Blow Back. Proposed modifications simply change MCAS settings to try and prevent entering this situation.Boeing’s explanation refutedBoeing blames the Angle of Attack sensors on PK-LQP, the Lionair 737MAX jet which crashed. Boeing’s theory however is easily disproven.It can be shown that the Lionair jet suffered a series of failures on 27 October 2018 flying JT776 Denpassar to Manado, (dates in images are just when posted)Then later the problem re-appeared on JT2749, Manado to Tianjin.Upon return from Tianjin as JT2748 the problem appeared resolved. Next date 28 October the problem repeated flying as JT775.After JT775 landed from a flight on October 28, the aircraft PK-LQP was removed from service and inspected. The suspect Left Angle of Attack sensor was removed and replaced with a brand new AOA sensor.It emerges that although the Lion Air jet had two AOA sensors MCAS only acted on data from the sensor on the Captain’s side.These roller coaster ride problems resumed as it flew JT43. More inspections were conducted after JT43 and no fault was found. Then PK-LQP flew the fateful flight as JT610.What was the Cause?The dilemma for re-certification is whether the problem was actually caused by the AOA sensor or originated in another related system called the ADIRU or DFDAU which the AOA sensor is connected to?The Digital Flight Data Acquisition Unit (DFDAU) may have encoded the data signal incorrectly. The Air Data Inertial Reference Unit (ADIRU) may have measured the AOA sensor incorrectly, or either system may have been subject to voltage irregularity.If any of this was the case then Boeing’s software patch may address the wrong issue. This is why proper certification is vital and also why China and the EU are correct to demand tougher certification from the FAA.Proposed SolutionUS airline Pilots drew President Trump’s attention to loss of control in 737MAX aircraft flown in US airline service. It appears something influenced President Trump to intervene and somebody made the President understand what both Boeing and the FAA were in denial about.From the flight manual to automation, why pilots have complained about Boeing's 737 MAX 8The point missed by Boeing was that these wild rides with loss of control happened both before and after AOA sensors were replaced. Nobody has satisfactorily explained why problems before replacement of the AOA sensor persisted after installation of a new sensor?For this reason I urge caution in re-certification that Angle of Attack sensors may not be the primary cause.Furthermore some of the problems experienced on earlier flights could not have been the result of AOA sensors because the AOA sensor had no direct input to themFlap Position influences MCASWith JT610 we can observe once flaps were raised Stabilizer Trim became highly active. This was immediately followed by arrested rate of climb and loss of altitude. Pilots on JT610 managed to restore the rate of climb but they increased airspeed to achieve this.Lion Air pilot Capt Bhavye Suneja lowered flaps again for a few seconds and inadvertently stopped Stab Trim stopped pushing the nose down. When he raised flaps again the Runaway Stab Trim resumed.Every time Captain Suneja hand trimmed up to correct Stab Trim down, MCAS pushed the nose back down again 9 seconds later.Neither pilot were ever trained to recognize the effects of MCAS. The crew became hung up on trying to identify what the problem was in their QRH handbook not realizing their manual had no advice about MCAS. In fact at that time Boeing had not even informed any airlines about MCASCockpit Voice Recorder (CVR) reveals the Captain calmly flew the aircraft hand trimming the nose up again each time against the Runaway Stabilizer Trim. His First officer consulted the Quick Reference Handbook (QRH), but could not find relevant advice how to cope with the problem.The Quick Reference Handbook (QRH) is organized by identifying symptoms and then leading pilots through a multiple choice menu to correctly identify a problem, then advise how to resolve that problem.Then frustrated when the First Officer could find the problem in their QRH, Captain Suneja asked his co-pilot to fly the plane whilst he looked through the QRH himself. The problem was that his First Officer never fully trimmed the nose back up but only partially did so.Because of this failure eventually air pressure overcame strength of actuators. JT610 violently pitched over. During the dive the aircraft went past vertical to become inverted. Both Pilots pulled frantically exerting over 100kg of force on their control yoke. Even with both pulling they could not raise the stabilizer again. This rare effect is called Elevator Blow back.Boeing never included any advice to the pilot’s QRH manual on MCAS.How to disconnect a runaway stabilizer was listed in their QRH however their Stab Trim was not behaving like a normal Stab Trim failure and this may have been cause of confusion. They had no idea that stopping the problem required a second step to disconnect MCAS.Neither their QRH nor FCOM mentioned MCAS. The problem kept resurrecting itself every 9 seconds unless MCAS was disconnected.Post Script:Originally, being in the dark along with everyone else I proposed different scenarios from those cited above. I interpreted lowering flaps intervening to halt runaway Stabilizer Trim as having triggered MCAS by disconnecting the autopilot. I am now advised the autopilot was not connected in the climb, so if this advice is accurate then I accept that.My theory of Autopilot disconnect when flaps raised was challenged in the replies below by Michael Groszek and I am willing to remove my theory, but meantime I have asked Michael for more information. This is what he had to say and I think it worth quoting him:Pontifications: Fluid, dynamic events upend MAX story - Leeham News and AnalysisI recommend reading the link Michael provided above.Other SpeculationsI also proposed back in November 2018 when the same thing happened on Flight JT43 that because it coincided with a PCU failure alert for elevators, which warns pilots of differential pressure on elevators, that perhaps MCAS recognized this as a jammed elevator?I suggested that this may have prompted spoiler deployment during the climb as illustrated below, killing lift:One of the tools which MCAS uses to counter a jammed elevator is to raise spoilers on top of the wing. These counteract a climb and dump lift.Pilots increased speed to counter stallThus if JT610 pilots were trying to maintain a climb or even just maintain altitude the MCAS system was constantly fighting them to descend the plane. JT610 pilots fought this by increasing to maximum speed. By increasing speed pilots inadvertently encouraged elevator Blow Back.Unfortunately given a barrage of stall warnings in the cockpit without accurate airspeed data maintaining high speed was a perfectly reasonable response.Similarity with Ethiopian ET302 is immediately obvious from both aircraft losing rate of climb.Proposed changes for re-certificationAccording to Vice President of Product Development and Product Strategy Mike Sinnett Boeing has developed an improved version of MCAS since the Lion Air crash.The key feature of Boeing’s changes is to prevent MCAS commanding recurring nose down Trim. Pilots were told at a briefing to pilots at Seattle on 27 March 2019, by Boeing, that MCAS which forces the nose downwards to avoid a stall, or loss of lift - will now only operate one time for each event rather than impose repeated corrections.Boeing have also decided to follow Airbus procedure and if AOA sensors are in disagreement by more than 5.5 degrees, in manual flight with flaps retracted MCAS will be disabled by the flight computer. So far the new system has undergone two test flights for FAA certification.Boeing’s responseWhat is of concern about Boeing is their management’s philosophy of denial & evasion until forced to act by others. This is not a culture of safety. It is one thing to fix a problem under massive public scrutiny and pressure by customer airlines and groundings by foreign aviation authorities.What Boeing lack is a culture of safety. The company is ruled by sales managers and accountants. Boeing has lost touch with aviation.I would whole heartedly support the current CEO being replaced by Captain Sully Sullenberger, famous for his ditching in the Hudson River of Cactus 1549. Sully is a through and through aviation minded guy. Most importantly he can be trustedBREAKING NEWS:Bjorn’s Corner: Why did Ethiopian Airlines ET302 and Lion Air JT610 crash? - Leeham News and AnalysisLatest development is the FAA is now under Federal investigation accused of encouraging Boeing to certify itself and also analyze the cause of LionAir crash with suggestions that the FAA abdicated its statutory duties.Report: Self-Regulation of Boeing 737 MAX May Have Lead to Major FlawsThe missteps that may have made the 737 Max crash-proneSince this was written it came to light the FAA had known of but ignored a database of reports by US airline pilots about control problems with the 737MAXBoeing 737 Max 8 planes: Were red flags ignored?Here is a link to loss of control problems with 737MAX reported by US pilots which FAA ignored after the Lionair crash:ASRS Reports for 737 max8FAA published Boeing’s proposed software fix in the CANIC report of 11 March 2019, but same report it notes that FAA proposes the fix without understanding what went wrong with JT610 or ET302. In other words, so far everyone is just guessing how to fix it. The FAA suggest the MAX will be back in airline service in May 2019.Breaking News update16 March report by Seattle Times that elevator jackscrew from ET302 found in same fully extended position as witnessed on Lionair JT610.alsoLion Air flight (JT43?) before JT610 crash successfully disconnected MCAS to avert disasterA day before it crashed, Lion Air's Boeing plane was reportedly saved by an off-duty pilotFurther reading:Youtube documentary on QualityControl problems at BoeingSimon Gunson's answer to Why has the US been preventing South and North Korea unification for the past 65 years?Simon Gunson's answer to What caused a brand new Ethiopian 737MAX to crash?Simon Gunson's answer to What is wrong with Boeing 737 Max Aircraft?Simon Gunson's answer to How are the engine position and center of gravity different on a Boeing 737 Max aircraft as compared to earlier 737 models?I also wrote this not long after the crash of LionAir JT610:Simon Gunson's answer to Is it true that the Boeing 737 Max is highly unsafe and risk riddled but its manufacturer is reluctant to call it back owing to the huge cost considerations?Why Boeing’s design department are responsible for loss of Malaysian flight MH370 through electrical failure causing hypoxia:Simon Gunson's answer to What happened to Flight 370 Malaysia from the pilot’s perspective?Simon Gunson's answer to Where is MH370, four years on?Simon Gunson's answer to Who is responsible for the Colombo serial blasts in Sri Lanka?

Probably. After all, most ventures fail. Tech investors, like bidders at an auction, are subject to the Winner’s Curse: those who err on the side of too much optimism are willing to pay more than those who calculate the risk/reward ratio correctly.From what I can glean from the BI article, the play here is to better evaluate the chances that a drug candidate will ultimately prove safe and effective. Given that the costs of failure can run from tens to hundreds of millions of dollars, even a modest ability to recognize and eliminate losers from the development pipeline would be very valuable indeed.Of course, this is not a new problem. There are thousands of successful drugs and many more thousands of unsuccessful ones, and drug makers have developed a number of “rules” that distinguish winners from losersFrom https://www.archie-west.ac.uk/projects/molecular-simulation-of-biological-systems/fast-and-accurate-predictions-of-physical-chemical-properties-of-drug-like-molecules/Lipinski’s Rule is the most famous of these, stating that orally active drugs violate no more than one of the following criteria:No more than 5 hydrogen-bond donorsNo more than 10 hydrogen-bond acceptorsMolecular mass < 500 DaA level of fat solubility (expressed as log octanol-water partition coefficient) less than 5There are many variants and exceptions to this rule, but you get the idea.Of course, incumbent drug companies are not blind to the notion that computerized analysis can give them an edge in identifying the best drug candidates. One report indicates that 18 of them have ongoing AI programs. Is BenevolentAI (the company in question) going to beat them all? Maybe. Is that a good bet at a $2B valuation? Seems unlikely.And how plausible is the underlying value proposition? The cost of drug development keeps going up logarithmically, and has for decades, a phenomenon known as Eroom’s Law (Moore’s Law in reverse).From Eroom's LawDrug discovery technology is light-years more advanced now than it was 50 years ago; a number of technological revolutions have been implemented. But the cost of discovering drugs just keeps shooting up. This suggests to me that technology is not the limiting factor here, and thus better technology is unlikely to be the solution.If there were a derivatives market in bad investments in biotech, I would go short on this one. But I would do the same with nearly all biotech investments. It’s a very tough game.