mswen

I had a grad school mentor William Wells who taught us something similar. A good research publication or presentation should aim for "just the right amount of surprise".

Too much surprise and the scientific audience will dismiss you out of hand. How could you be right while all the prior research is dead wrong?

Conversely, too little surprise and the reader / listener will yawn and say but of course we all know this. You are just repeating standard knowledge in the field.

Despite the impact on audience reception we tend to believe that most fields would benefit from robust replication studies and the researchers shouldn't be penalized for confirming the well known.

And, sometimes there really is paradigm breaking research and common knowledge is eventually demonstrated to be very wrong. But often the initial researchers face years or decades of rejection.

A friend and I built a proof-of-concept of using a variation of Latent Semantic Analysis to automatically build up conceptual maps and loadings of individual words against the latent conceptual vectors back in 2000. In exploring what it would take to scale I concluded, like you, that we should use professionally written and edited content like books, news articles and scientific journals as the corpus against which to build up the core knowledge graph.

Twenty-four years later I still regret not being able to raise money to enable us to keep working on that nascent startup. In most ways it was still too early. Google was still burning through VC money at that point and the midwestern investors we had some access to didn't get it. And, honestly they were probably correct. Compute power was still too expensive and quality data sources like published text were mostly locked up and generally not available to harvest.

I recall in the late 70s that my high school also had a teletype terminal and an IBM card reader that connected to a mainframe for the whole school district. As a student I had some awareness that it was unusual. I was also working PT at a Radio Shack at the same time and saw the first arrival of a TRS-80 to our store.

Despite that early exposure to computing technology I went other directions for the next couple decades.

Different recession. Spring of 2008 I was in a corporate job within a fairly new business unit. Over the space of 3 years we had grown the new business unit from nothing to a unit with about 25 employees and turning a profit. For the first few months our business unit wasn't touched. We saw other more mature business units in the corporation going through selective layoffs to tighten up expenses. But when the recession started impacting our revenue growth, from a 50 to 60% annual growth down to 5% growth the layoffs came to our unit.

What I provided to the business was one of the ancillary services rather than the core service offering. So regardless of the good performance reviews when the word came down that our unit needed to cut costs the only place to cut was labor and I and a couple other people got laid off in the first round. In many ways I couldn't fault the decision.

Oddly enough they laid off the sales person who was most responsible for their growth. He had the misfortune of having been given the challenge of opening up a new market just as the first of the recession was being felt in 2007. Honestly I was more surprised at his layoff than my own. But I guess whoever was higher up in the corporation directing our layoffs was more focused on what have you done for us in the past 3 to 6 months than what have you done for us in the past 3 to 4 years.

This reminded me of a conversation I had with a business owner in southern Florida. I was showing him a demo of some mapping technology I had played around with and mentioned that I was also involved in some web crawling and NLP stuff. He mentioned the following opportunity.

If you could crawl and scrape public records to determine when the Air/Conditioning units had last been replaced at each property and then map it. He was convinced that HVAC companies have a good sense of the age of units that are needing replacement. One would generate lead maps specific to properties so that a sales person could focus only on the properties that had units approaching replacement age. This was around 2010. I poked around in the public records a bit and decided I wasn't the right person - even if technically possible.

I am not sure of the market validity of this idea but you might be better suited to this kind of thing than I was.

I lift Monday through Friday at about 11 AM each day. One week I will do pressing exercises such as bench press, squat and leg press on MWF and pulling exercises such as deadlift, hamstring curls, lat pull downs on TTH. Then the next week I reverse the days. That way every major muscle group gets 5 workouts over a 2 week period.

I do a few minutes on an exercise bike to warm up before lifting and usually finish off with another few minutes on bike or high intensity sprint on a rowing machine.

I don't eat anything in the day until after that workout. If I am trying to lean out I will extend the no eating until supper and just eat one meal.

Weightlifting as a consistent form of exercise. I love to run but I have had issues with my feet. With weightlifting I have been able to work around any injuries and keep consistent. I started in my mid 40s after a foot injury that ended a good couple years of running and bodyweight resistance. When I couldn't run I started lifting much more consistently and actively building strength.

Lower back pain, that I thought just went with middle age went away as I got stronger. Although I had one weightlifting injury that caused me to cut way back for 6 months I eventually recovered and went on to be stronger than before. I am now 61 and I really notice the difference as I look around at other men my age. Seventeen years of consistent resistance exercise makes a big difference.

My son is at the other end of the spectrum having just graduated with a BS in Computer Engineering and started his career as Software Engineer I. He is working as a software engineer at a company that designs and manufacturer's sensors used in factory and warehouse automation. Creating the software to do interesting things with sensors is a growth area for them, but it is still primarily a hardware company.

If that kind of thing sounds interesting let me know.

Start-up 1 (2000): Too early and inadequate access to technically sophisticated investors (located in the Midwest rather than Silicon Valley). Proof of concept in semantic knowledge graph but would have required computational resources to do it at scale that very few were building at the time. And, cloud wasn't a reality at that point. VCs started funding "semantic" about 7 or 8 years later.

Start-up 2 (2010) I was employee 7, not founder. It got through series A funding. We built it. We won award for innovation from Sloan School of Business and although CIOs of big companies professed to love the platform when they passed it off to their procurement and vendor management people adoption faltered. And, without a significant flow of RFPs being posted there was no revenue as we were going to get paid with commissions from the vendor side of the 2 sided market.

Start-up 3 (2012) - Solo founder builder. Lack of market fit. No funding, needed to earn a living so it went on the back shelf. Some technology still used for consulting clients but not as a product.

It all depends.... if we were having a private conversation I would be asking about all of the following. But I wouldn't really recommend getting too specific about some of these on a public forum.

How old?

Current health status? Chronic illnesses? Cognitive health?

Current level of independence or dependence for daily living?

Recognition of need for assistance?

Level of wealth, property, investments and liquid cash?

Level of income - degree still dependent on parents working FT, PT or now on pension, social security or other retirement monthly income?

Level of debts needing to be paid off or serviced?

Presence of non-family social support?

Living proximity to you?

How much time monthly can you devout to helping?

What is your own financial strength and ability / willingness to help financially?

What level of trust do your parents have in you?

How comfortable are your parents with technology?

Do your parents have access to high speed internet?

Availability of assisted living and/or nursing home facilities

And more

I personally have walked a pretty long road of being the lead caregiver, guardian during 20 years of declining health and increased dependence from my own parents with each eventually dying in their mid-80s.

It feels like the only generic advice is something like the following.

Spend regular time with them and pay attention to changes

Keep their dignity in mind - no one likes the process of becoming more dependent on others

Be willing to calmly but firmly advocate for them with doctors and institutional caregivers

Technology may not help nearly as much as you imagine

Effective assistance will be more expensive in time and money than you currently envision

I am not sure if Jason Roberts was the first to use this phrase but here was his take on Luck Surface Area back in 2010. That was the first I personally heard of the term Luck Surface Area. I don't know that Jason's take is any more insightful than the linked article, but I feel like he deserved a call out as well on this topic.

https://www.codusoperandi.com/posts/increasing-your-luck-sur...

I went to buy something on eBay and found that my account had been suspended. I have never sold anything on eBay. However, I had signed up for an eBay developer account and then never used it because the client who I was exploring it for went another direction. So I thought maybe it was related to the unused developer account. The support person couldn't really tell me anything but said I could contact some part of support for an appeal so I asked that they send me an email with that process. They said yes, our chat is automatically emailed to me. But no email followed up that support experience.

Very poor support. No explanation and action including actions that are promised on their support page.

Oh, I totally agree with you. And, there are times when it is not useful to try to come up with a solution because the other person just needs to be heard. It is not really about problem X. The real issue is not feeling heard, respected, loved.

Your comment reminds me of the following. My wife and I have been married over 30 years now. Our total household is 7 persons.

A couple years ago, my wife was complaining once again about someone using scissors and not bringing them back to their proper storage place. "How can we have 3 pair of scissors and none of them are here when I need to use one?" This didn't bother me but hearing her complain about it did bother me. After a couple attempts to reason, "it isn't that big of deal to track a pair down" or "how often do we really use them?", I decided that abundance was a better solution. I found a 4 pack of decent scissors for about $12.

So for $12 dollars I have never heard that complaint again because even if someone walks off with one and doesn't get it back right away there are several more. So my wife doesn't doesn't experience that frustration and it keeps her from getting fixated on something as insignificant as the location of pair of scissors. And, I have already decided that if it happens again I will buy another pack. They are surprisingly good scissors for $3 each.

I think my broader point was that we as humans are sometimes irrational about certain annoyances in life. And, if I can find a way to spend some money and just solve the issue that is probably a good use of money.

Some great points AlDante2. There is also the issue of how open-ended the engagement is versus clear confined specification.

I work with clients on an hourly basis because of the open-ended nature of the work I generally do. It is almost impossible to write a decent spec ahead of time. I do a weekly check-in with demos and discussion so they know what I am working on, how much progress is being made and with weekly check-ins they don't have any surprises at the end of the month when I bill them.

Recently, I added a new client where I was much faster than the estimated hours for the initial scope of work. If I had been fixed bid with them I would have made a lot more per hour. As it is they are using the remainder of the hours in the original estimate for some other work and it sounds like there will likely be more in the pipeline. Truthfully, I would rather have them turn into a longer-term client in a high trust relationship than to have maximized profit on the first project.

One of my good friends spent a couple of years selling grain silos to farmers in the American Midwest. I heard about this some years later when we were living in the same community again and able to get together for coffee once in awhile.

He commented how often the following question opened farmers up to a much deeper conversation that had nothing to do with grain silos. His opening question would often be "How goes the battle?"

Although he did well at selling, he is a person who is genuinely interested in others in a warm caring way. So I am sure it was his manner as well as the specific question. But I love the question as well.

How goes the battle? Immediately acknowledges that life isn't easy, that the things we struggle with are okay to talk about. However, if the person isn't in the mood to go deeper they can just respond with "fine", "alright" or "can't complain" and just move on without going deeper.

But asked in a genuine caring way, "how goes the battle?", invites a deeper more authentic response about struggles.

Trustworthy - they will have access to sensitive information.

Articulate - they will communicate on your behalf.

A good dose of common sense.

Comfortable in the role of follower but not a pushover, they will be helping maintain your boundaries.

Willing to do mundane repetitive tasks, content in the knowledge that they are freeing you up.

Consistent, calm and predictable.

Process oriented - help keep things working smoothly.

The distinctions between self-control and grit as articulated in this paper speak to one of the fundamental issues that start-up founders face. Defining a superordinate goal that will be pursued with grit, passion, zeal and persistence over a period of years or decades. Versus the self-control at a more granular level which has much more to do with productivity.

But the key is what to do when something (usually market fit) isn't working. Do I persist and pursue alternatives but still all directed toward meeting the same superordinate goal? How long do we persist? How many alternatives do we explore? At what point do I determine that a pivot is needed? That merely changing a lower level approach is failing and I need to take these resources including core team and deploy them to a different superordinate goal?

How do you go about deciding when persistence in the face of failure is enough and it is time to change the superordinate goal?

From this comment it seems you are intending to pursue the masters while working. So the opportunity cost is not lost salary. Your opportunity cost would be measured more in time and stress. You will be giving up social time. If you are single and don't have children and are young enough that the masters degree has time to payoff, then I would encourage you to go for it.

Also, since you really enjoy both theory and hands on coding you are well situated to get maximum benefit by applying theory to real problems.

You are correct that this is very difficult to untangle. But remember features that initially excite and please customers because they were new and provided even unexpected benefits, soon migrate into "more is better" kinds of features and then over time every competitor is offering the same feature and it is "table stakes." You really only expect sales upticks for features relatively early in that lifecycle. The rest just contribute to customer retention.

You should assume that most businesses already have some kind of solution in place, either from a competitor or an internally developed a set of processes and supporting software. When you look for sales bumps you need to consider the length of the sales cycle - you might not see any switching for 2 or 3 quarters after it becomes generally known that the feature is available.

I think your surest sign is in customer retention statistics. If normally you experience 10% non-renewals of license every quarter. I would look in the quarters right after it is introduced to see whether renewal rates improve.

And, still your critics might say well it wasn't the new feature that your department created and introduced, it was the new marketing campaign and we revamped our process for renewing customers. And, you know they might be right. Or, the improvement is a subtle combination of all those factors. Unlike science, businesses, particularly B2B, rarely do controlled experiments.

You can track inputs and outcomes over time and point to correlations but that is about the best you can do.

In an organization that size there should be a whole team of internal experts as well as independent, external customer and product research resources devoted to understanding customer needs and desires. Those needs should be researched and matched up to projected solutions within your product.

Those high level solutions should be articulated in a feature roadmap. If appropriate product research has been conducted you will have explored the problem and potential features in depth with customer interviews, focus groups and broader survey customer research to gauge impact on customer retention, license upgrades and then extended that research to understand impact on acquiring new customers.

As an engineer with a feature idea find a way to get it on the radar of that product research team so they can see about validating it with customers.

Having been a professional researcher doing exactly that kind of validation at one point, and now doing a fair amount of development, I have come to believe that building a sandboxed prototype is better than just trying to describe the conceptualized feature. But be careful not to invest too much, time, money or emotional commitment in the prototype. Get it in front of customers for reaction as soon as possible.

If you want more examples and stories feel free to reach out.

I am surprised by the results of this study. My own experience is that Time-Restricted Eating helps with weight loss when I decide to lean out. I tighten down the eating window. And, I lose weight without getting very strict on what I eat and I don't feel deprived.

I am thinking that it mostly works by just making it easier to eat fewer calories over the course of a week. When I was going through a a year plus when I was working on max strength weight lifting I could eat enough in my 8 hour window to gain muscle (and some fat) and reached my lifting goal.

When I had to skip the gym for awhile this spring due to COVID-19 I lost some of that muscle but kept eating (maybe due to stress) so I put on a couple more pounds of just fat. June first I decided to make a change and lean out. So I started restricting eating to 3 hours and sometimes down to 1 hour (OMAD). I did this M-F and gave myself freedom to eat additional meals on the weekend.

The weight immediately started dropping off. Then late June I started jogging and reduced weight lifting down to a maintenance level. I have continued all summer and lost about 7% of body weight - some muscle but mostly fat. I will likely continue this routine for awhile focusing on getting substantially leaner.

I have followed the research literature on TRE, IF, ADF and prolonged fasting, and read many personal accounts like mine that suggest TRE is an important component of overall metabolic and weight management. If this study stands up then it would seem that most of the observed effect of TRE is because it helps people eat less without feeling deprived. And, clearly it is easier and more approachable than prolonged fasting.

It will be interesting to read responses from other researchers in this space.

A family story. My father-in-law (now in his 80s) began to stutter when he was very young, somewhere around 4 years old. Teachers, and I think his step-mother were trying to force him to write with his right hand, even though he naturally gravitated to using his left hand. At some point while still a child they gave up on trying to get him to write with his right hand and shortly thereafter the stuttering went away.

It might have been anxiety caused by the forced switch in handedness. But if so, why didn't stuttering reappear at other stressful times in life?

Or, is there a neurological connection. Writing and speaking both access language centers in the brain, so trying to force unnatural change in writing created problems in the oral expression of language?

Does Kipsu do internships? I have a son who is starting his junior year at U of Minnesota, Twin Cities in the Computer Engineering program. This past summer he had applied for a bunch of internships and with COVID-19 shutdowns everything just dried up.

Changeling (The Dark Fae Series Book 1)

Faery Tales (The Dark Fae Series Book 2)

By McKenzie Rae

Book 2 was just released last week. I sat down this morning to read and as of this evening I am about 200 pages in. It really sucked me in.

Disclaimer. I am the father of the author. But when she was younger I would read something she wrote and if she asked for my reaction I was honest. I would say, it is really good for a 14 year old. That would drive her crazy. But I would tell her I will always give you my honest reaction.

When she was picked up by a publisher in her early 20s, she was kind of bashful about me reading her work. It was so gratifying to be able to walk in about 2/3 of the way through that first published novel and tell her that even though it wasn't my preferred genre I had started to care about the characters and the plot and I had to finish.

I feel like each book has gotten better. This latest book captured my attention within the first few pages and I spent way more time reading it today than I originally intended.

Sorry if anyone is offended at me promoting my daughter's work.

I went down the rabbit hole of looking at Idyll [0] the tool that he uses to create the shuffling simulation. Some pretty interesting examples in the gallery. Trying to decide whether the capabilities are worth the learning curve.

[0]https://idyll-lang.org/

Totally agree with you. Oddly enough I am currently doing a market research project on commercial applications for Quadratic Unconstrained Binary Optimization (QUBO) and also Constrained Binary Optimization. Anyway we were talking about UPS, and just for reference if I search my 3 degree network on LinkedIn for the combination of CPLEX (an IBM optimization package) and UPS I got almost 400 hits. Now admittedly some of those people used to work at UPS, some work there now. But there are clearly many, many computer scientists, operations research scientists and mathematicians working on these and other optimization problems at UPS.

On the other hand the grandparent comment just illustrates how even with all those optimization resources individual "decisions" may still be stupid in the moment even though the overall solution is "good enough".

Agree they are using this context for eigenvector without going into detail. Another way I have used PCA and/or cluster analysis as an analogy with a group is to get them to put all the questions, product or service features, or customer benefits on large sticky notes. Each person generates as many as come to mind in a 5 to 10 minute period.

Then taking turns each person gets up and places a couple of the questions or features they have generated on a wall or whiteboard, briefly explaining what they mean while also placing it relatively near or far in conceptual space from the items that are already up there.

As you get further along, you invite people to start reorganizing the sticky notes talking about why they think some of these questions are more or less similar to others. As they explain their reasoning you start hearing the underlying latent dimensions, the organizing vectors.

When the process is all finished you have what looks like a cluster analysis. Then as a group we name those clusters. This qualitative dimension reduction exercise often has the result of getting a group of people talking about the question or feature space in a common set of reduced labels and gives them a common framework.

I found these techniques very enlightening to use with product groups as we prepared to use research techniques to capture a Voice of the Customer.

I am reminded of a conversation with the CIO/CTO of an Au Pair matching service.

Their process at that time (2008) was to have a set of employees who were matchmakers. They would read hundreds/thousands of family requests and requirements and then hundreds/thousands of applications of the young people from various countries around the world. They would match them with prosperous US families that were looking for an au pair.

As he explained the process I began brainstorming ways to use NLP and earlier versions statistical analysis, machine learning or maybe genetic algorithms to take over or at least assist in what was a very manual process by those matchmaker employees.

He left for a different job not too long after and we never did more than a bit of brainstorming. And, I suspect that this matching is still being done largely through human reading and thinking including how humans "read between the lines" which can sometimes be so accurate and other times just reveal the biases of the matchmaker.

However, thinking about those kind of person to person compatibility and matching problems led me to reserve the domain name trupeer.me - the idea of helping professional career people find their tru peers in industry

San Diego CA for that So Cal laid back beach atmosphere plus a good tech scene if you are or can be oriented toward biotech.

Santa Fe NM - feels like it has roots. Unique culture and great food. National Labs - Sandia and Los Alamos not too far away for a good science flavoring.

Boulder CO - great outdoor recreation opportunities, university and research institutes support science and tech vibe