Monday, May 14, 2018

Arbitrary

anyone with permission is granted that...
anyone...


Reducing image noise for nighttime shots

Machine Learning enhancing night images

Noisy night images are definitely headed out the door.  AI and machine learning will likely help users produce clearer and higher quality productions without so much user effort.


Sunday, May 6, 2018

Op-ed to op-ed: After 14 years...

Here's the original article:

After 14 years, I’ve had it. I’m leaving Seattle

Honestly hadn't lived there for well over a decade, and that being said, had few to any run ins with the homeless even in areas of the city where populations had higher numbers.

Involuntary treatment actually is not customary in so far as mental health treatment around the country and there is a reason why this is the case.  One because of the historical legacy of forcing people into treatment and the legacy of abuse, the idea of going back to forcible treatment is probably not a popular one.  I'd suggest to the science writer to do his research on this topic.

Alex complains about spiraling costs of housing...well, Seattle trends like any other higher demand West Coast city that is likely going to have increasing housing costs.  A city that has increasing international real estate appeal is one that is also more high demand and likely becomes less affordable to locals that aren't as wealthy.  Get over it, and find another place to live if its too much.  A lot make their money, get their job experience,  and leave to other parts of the country where the cost of living is more reasonable, and where certainly housing is more affordable.  It may not be as pretty scenery wise elsewhere but you may find in the long run, you have more money in your wallet for travel anyways.

Decided to move to the East side, Bellevue, Issaquah, maybe Mercer Island, Kirkland, or wherever else? 

Complains about minimum wage hikes...
Gentrification of the city hadn't helped, but honestly there were a lot of locals moving out of Seattle city proper a long time before you came along.  Seattle has mostly had a narrative of revolving people in and out...entire building in Fremont (on the historic registry) moved to Ballard to make way for Whole Foods now defunct.  The city has continually faced erosion and a lot of migratory transplants for quite some time.  There's nothing new in your op ed here.  The wealthy want their businesses to serve them, let them have it, and expect wage standards advocated.  Honestly I've traveled to countries elsewhere where food costs are exorbitantly higher (double to quadruple) what typical American costs are...such country appears to be managing quite well...business still operate there, and people are still willing to pay for the costs (tourists and locals alike).

So what if you leave...I did...like a lot of others in the revolving door of the modern American landscape.  Try picking a place that is a little less desirable and maybe you'll find more civility.


Tuesday, March 27, 2018

The rigged imager

   Recently having purchased smartphone X from store.  M__ attempted numerous self portrait shots.  Numerous more than he could say, and all images seemed to provide a distinction relative to the image that were commonly viewed in front of the mirror.  Why is it always very similar?

Downcast and disheveled face, cheeks puffy and slightly discolored, M__ appeared tired and worn, he appeared haggard.  Obviously outdoor lighting seemed to provide greater improvement for his appearance.  Camera's don't lie, do they? 

The truth of the matter for a ccd imager an image is processed and calibrated in some representational way for all the incoming light, the imager is calibrated in terms of human visual perception.  It isn't accurate in some ways as it is calibrated in an average way to visual perception.  Some see colors differently, and some perceive the vibrance of some colors distinctly relative others...and some even see colors in sounds, and smells, as connected to other senses, that is, given a neurological crossing of wires mixing sensory states.

The truth of the matter is that even a compiled image from a smartphone or camera is likely to be processed whether a user likes it or not.  That is reprocessing an image potentially from a RAW digital format into other media formats which approximate original raw binary data through a number of mathematical transformations, and not permitting all the calibrations that supposedly are meant in emulating a perception like human optics.  The image is filtered, and so it is assumed that the image in a given instantaneous moment that is still image is gathered is representational of its subject matter, save the apparent snow that has gathered in the image...could this be the interference, for instance, of something like cosmic rays having perpetuated when not enough visible light should exist for the duration that a shutter had gathered whatever photons that could be had? 

The light were more sickly yellow than he would have recalled, at least, if it were his eyes, neither had he perceived so much white noise (snow) in the image of a room that appeared darker than he recalled, having shown him in unflattering ways. 

A clinical white wash of light shown overhead as M__ gazed into a mirror.  Yes, perhaps, I can see more a blemish here and there.  Signs of aging that he'd convince himself must have existed only having overlooked this or that, the camera hadn't perpetual misheard him, but looked upon the details in some objective way it would seem...don't we have a tendency to think of ourselves in continuity to a greater depth than extends beyond surface?

M__ noticed over the course of months and years that his face changed, as one might expect with years in aging.  Changing as in some mutability of self, that one would expect, irrespective of the continuity of the self "I" that M__ knew.  I being there, M__ thought. 
A slight shift in angle, a distinct head posture, a facial expression that essentially conveys something beyond a transient misappropriation of body languages that coincided for all purposes to bad timing, but just so happened to be the case more often than not.  So much less thought into that resident expression that betrays all other expressions. 

There is no grudge in that discrimination given by algorithms.  It is posited as certainly as facial recognition employs the framing of a face and has a spatial appropriation created so instantaneous that the tag box need be employed perpetually in the frame() method (until instructed otherwise).  That is where the eyes, nose, and mouth exists.  The curvature of that mouth alongside the topological models that would be instantaneously conceived following the smooth curvature of facial muscular structures.  A lifting of the cheeks, lifts the corners and creases  of the mouths...long since understood by those in art and medicine.  Turn that smile into a frown.  Cross one eye with another...a strange asymmetrical gait is more noticeable than one that is not.  Human minds by evolution discern such noticeable patterns right away, and in sensing discriminate one face from another.   Machine language manipulated all such data in a dizzying way.  Much more rapidly than any human artist could conceive, that is, in continuity of one thirtieth of a frame per second or 1/30 seconds a rendered frame with all necessary ingredients to convey what need be conveyed.

M__ wouldn't have better on a given day that he were apparently tired, wouldn't have felt better on a day that seemed haplessly the same as the day before. 

For the endless string of days, he were postured like a puppet in an endless choreography, all such social capital spent away, and where likely social credit should be less common place.  Restricting his travel was all part of the system, as in the litany and narrative, describing its own self continuity, to readers and onlookers so often far away. 

His mind traversed the image that would be his own self memory, self invention that surely was in his own domain as equally as it were removed.  Something that no one would see in a world where few did gaze so much at the faces of others really.

Friday, March 2, 2018

Recent Study on the Ride Share Gig Economy and Added Thoughts

Here is the study:

http://ceepr.mit.edu/files/papers/2018-005-Brief.pdf

Notable highlight:

"We perform a detailed analysis of Uber and Lyft ride-hailing driver economics by pairing results from a survey of over 1100 drivers with detailed vehicle cost information. Results show that per hour worked, median profit from driving is $3.37/hour before taxes, and 74% of drivers earn less than the minimum wage in their state. 30% of drivers are actually losing money once vehicle expenses are included. On a per-mile basis, median gross driver revenue is $0.59/mile but vehicle operating expenses reduce real driver profit to a median of $0.29/mile. For tax purposes the $0.54/mile standard mileage deduction in 2016 means that nearly half of drivers can declare a loss on their taxes. If drivers are fully able to capitalize on these losses for tax purposes, 73.5% of an estimated U.S. market $4.8B in annual ride-hailing driver profit is untaxed."

And this is the rub for an industry that maintains a substantial market.

Added thoughts:

Uber or Lyft having externalized transportation cost infrastructure to ride share drivers removes incentive in provisioning scarcity to ride share driver supply.  While on the other hand, potentially encouraging market growth for demand, as some consumer have related, the average time to service could be in some cases considerably lower relative to traditional cab services.   Market saturation of supply drivers likely leads more likely to shortened arrival times for supply customers while ensuring greater likelihood of ride share orders being fulfilled as opposed to cancelled.  Also having externalized transportation costs in such way, it has the advantage in cost leveraging the cost per ride to consumer while maintaining profit margins relative to traditional ride services.  This displacement, of course, is passed to drivers of the ride share service.

Here are additional factors that can make pay rates quite low:

  • Cost per ride doesn't pay as much when minimum hourly pay rates are non existent.  
  • The frequency of rides per hour are small alongside small mileage added to lengthier deployment times and mileage in providing a ride in the first place.  That is, unpaid travel distance meeting small paid travel distance.  
  • Too much downtime travel distance.
  • Too many drivers in queue for a given location (e.g., waiting at the airport with 96 others in queue), not enough demand.
Maximizing travel fares:
  • Frequenting locations where clients may be utilizing transportation in specialized ways as opposed to a primary source of transportation.     
  • Choosing locations where clients that use ride share as primary means more frequently may be more likely to use the service, in absence to public transportation offering, for shorter trips and especially in non specialized ways.  
  • Events based ride shares could potentially have better pay outs since these are more likely to require highway miles and lengthier travel times.  
  • Pickups with drop off at the airport especially where travel distance to and from are increased. 
  • Reducing non paid travel time.
  • Finding minimum distance routes to higher frequency ride share demand locations.
  • Use multiple ride shares (if possible). 
  • Reducing total per day travel miles while increasing the number of rides per day.  
  • Offering ride share when the pay makes sense.  What is a base hour rate goal?  Getting paid to do other things when typical base hour rate during such time isn't in keeping.  
  • Doing ride share in conjunction with other types of paid services (e.g., not Uber delivers but other types of specialty delivery services that pay decent).  Thus signing on to Uber to offset downtime loss of income when other delivery and/or transport services are slow as means to supplement primary income as opposed to primary means.   
Because Ride Share conceptualizes the market of drivers as intelligently driven by the supply of drivers and consumers more so, there is likely more managerial stress burden placed upon drivers to make critical economic decisions as to whether or not the market is viable as a means to income.  The gig economy is revolutionizing the ways that time is spent and the valuation for such.  Like the outset of the industrial revolution and all ramifications entailed by emerging technologies and sociological manifestation therein, it as likely that society is being transformed to think in different ways about the utilization of time and resources.  There is the power of potential exploitation and maybe in the future greater empowerment for individuals in maximizing their returns for time spent.  Proliferating diversification of gig related work is a likely reality for our economy.  Increasingly companies, corporations may be looking to pay independent contractors for task related work as opposed to lengthier paid downtime stays.  This inherently puts greater time management stress on individuals in such economy to maintain some paid work load when it is necessary, but also being intelligently cognizant in ways that were less commonplace in the past.  Knowledge and task based services as they become increasingly transient in terms of continued usage will likely mean less are as highly specialized in providing services and more likely having better adaptive management skills and/or utilizing adaptive management services that makes more likely effectiveness of individuals in being able to deliver profitably skills sets, services and products.  One should predict that our future economy will likely put greater premiums (not less) for task service related deliveries if it is ever to be sustainable.  Of course, externalizing lean efficiency is yet another thing... 

Thus a maxim:  Don't be afraid to express your worth for what you are doing and don't settle for less.

Tuesday, February 27, 2018

Icelandic Summer Landmannalaugar Solo Trekking and Camping Guide

   I want to make this post brief.  Mostly there's a lot of travel guides out that mention considerations to Icelandic weather, heed the advice, but I'll add something additional here.  Iceland is a Northerly country subject to more extreme weather conditions even in the summer.  It can be sunny and generally nice, even while it is cooler, and then it can be more extreme in terms of gales and stormy weather.

This past summer I managed to do the Landmannalaugar trek, but also encountered more severe gales.  The weather events were enough to have a warden shutting down the trails, and strongly suggesting camper's (with tents) to take up adequate shelter.  One particular camp was closed so that my starting day thru hike from the first destination point was instead differed to a third camp site.  That is going from the interior highlands south to the lowlands.  I decided (owing to getting ahead of an incoming storm) not to stay at a higher altitude alpine location and instead thru hike to the second location which was lower elevation, but even so were deferred to a third location.  That being said, extreme weather events can lead to camp closures if warden's make decisions that adequate shelter isn't afforded.  Also expect thru severe weather events that wardens may close the trails down (in our case a full day).  Quite likely though you wouldn't want to be out there anyways.  Just keep this in mind when factoring time schedules.

Fortunately the trek does provide a number of bus opportunities, but I would suggest taking note of schedule if you intend to use this.  Don't assume that huts (because they are advertised) are likely to be open.  First come, first serve?  Sort of, but not really.  During the height of tourist season, as I experienced, and especially during a severe weather event, the huts tend to be in high demand, but especially by paid tour groups, so unless you have exclusive reservation through a tour company, don't count on a hut being available, and make that even less likely if severe weather comes.  Going with paid touring companies affords more likely access to huts, and/or luxury treatments for trekking.
Wise to bring a three or four season tent.  I rode out severe storm with extremely high wind gales in a three season hiker's light weight tent which served well especially when sheltered in with a stone wall for weather reinforcement.

If it storms and severely so, you will likely deal with plenty of water (especially given all the river and stream crossings).  The deepest stream was nearer to the Thorsmark terminus which (at that time of the year...July) was waist high and generally a stronger current...I am six feet in height.  Highly recommend trekking poles for the crossing and/or managing to scout an adequate channel crossing site.  Keep pack belt un clipped so that your pack can be evacuated from your back in rapid fashion should you fall into the stream.  The water is glacial cold.  Wet shoes with decent traction and/or water sandals (same) can come in handy.  Gaiters are likely fine for most of the highlands, not so good where the streams in south are deeper...most stream crossing can do better with water wear for your feet.  Just change before and after crossing from dry to wet footwear and so forth.

I had witnessed a number on my trek through hiking to Thorsmark in a two or three day journey or busing out earlier because of weather conditions.  Most it appeared especially came into Thorsmark early (from our starting group) likely because of wet gear or the inability to keep gear sufficiently dry.  I, fortunately managed to keep gear dry through the duration of the severest weather, but found that on the tail end of the worst storm, my gear was saturated.   Thus I would suggest some added things for a tent.  One, it has a rain fly vestibule area where hiking boots and/or additional gear not stored in the tent can be kept in dry safe keeping, or that alternately you have a rain tarp situated so as to provide your tent additional dry space.  Secondly, compartmentalizing tent and sleeping bags or anything in tote from one another to ensure that water doesn't creep saturate internally other parts of your gear another thing to keep in mind.  Stowing an external rain tarp outside your pack can simplify this, or keeping the rain fly of your tent plastic bag stowed and segregated from other dry gear should help a lot.  Don't assume a rain cover for your pack is enough.

High humidity can more likely bring ruin to your down bag experience...consider synthetic bags.  Consider woolens and synthetic wear.  Bring and expect to use layers. 

Another small secret I'll share with you...I asked a warden, if my family home would see present weather conditions where I was at...she replied, 'That information most certainly will not be published outside Iceland!'  While I used free Wi Fi hotspots throughout Iceland.  I found none on the trek.  Something to be aware of if you intend to use wifi as a means of communication.  As to cellular service and using your existing phone abroad, check in advance.

Guides will tell you to expect much...don't assume that the weather reports that you see are what to expect before you go...its always sunny and nice, till you get there... who wants to lose revenue anyways?  :)


Saturday, February 3, 2018

React and Redux

What is the difference?

One I hadn't wanted to attempt as thorough researcher to answer this question with technical ins and outs, but rather approach this with a simplest possible answer, at least a reduced answer that avoids all the technical ins and outs of differences here.

One while react does provide in its framework controller model and controller logic interfacing, by far, React provides a framework with the design aspect of encapsulating mvc structure in a component by component basis and View predominant in its focus.  While React can be designed functional in nature, it seems to me to lend itself to classical OOP design patterns.  It is, of course, also quite easy to learn and trending popular in terms of adoption and use.

Redux apps on the other hand may share some structural difference to MVC design patterns in the way of separating Controller and Model logic with the that of the view systems, as opposed to having a compact all in one strictly component integration.  That is, in housing model and controller logic to the component.  Redux logic separate Controller logic into, for instance, actions and states of the app, and making ease in generalizing the state and actions of the application across components and containers, which are components that merely handle the delegation of component actions and states.  It is also one of a functional design pattern that provides injection patterns into component and containers alike, so it also extends functional design in a positive way given flexibility and power in evolving state and action of the app.  The downside (if such is appropriate) is that Redux has a higher learning curve for adoption, at least in the pure sense of learning redux without middleware helpers.

Which to use?

If you were wanting to gain experience with Redux design and especially functional programming work, it seems Redux would be a good choice.

If you have complex state and action managements for you Web Apps design patterns, Redux may also be  another good choice, though as I've read for asynchronous management, you may need some added middle ware helper (Redux provides advice on this).

If you are new to creating Web Apps or want something that easy to dive into, React is great as a learning tool and creating a web app quickly.   React provides great versatility and state management is especially easy to understand and implement.  However, complex state management systems with two way bindings and especially implementing a generalized state management system could be also just as cumbersome work in React relative to Redux.  Thus I've seen, for instance, WebRTC web apps designed in Redux relative to being purely React, and there's likely no coincidence where app states and actions would need be shared across a number of components since an overarching controller system provides some clarity to the approach of binding data.  Redux, add goes beyond what Angular provides here especially where Angular component models increasingly resemble the simplicity of React's components which has been the trend apparently for any number of web app frameworks.
 While two way bindings, for instance, attend to the problem of two way state mutations, in React or Angular for that matter, on the other, hand I've found myself violating the mutability principle of functional programming design.  That is, instead of mutating the inputs, providing a return object that expresses a new state.  Redux, in this way, could be considered the preferable approach if and wherever app design patterns make use of two way bindings between child and parent components and is required.

If you know neither, perhaps, it may help to understand React first and then learn Redux.  Though when learning Redux provides for enough distinction to think of Redux app by way of Redux design through its native api documentation (React may not be helpful) alone.  Trying to understand Redux, through React documentation, in other words, can get in the way, in my opinion.

React is to Redux, however, arguably for what Python is to C++...which is something to keep in mind. 

Oblivion

 Between the fascination of an upcoming pandemic ridden college football season, Taylor Swift, and Kim Kardashian, wildfires, crazier weathe...