Oh boy -- I'm the cofounder of Wanderlog (https://wanderlog.com), a YC W19 startup exactly focused on planning travel. And I've really read all these articles
But we're still at it and have grown a ton over the past year!
- On frequency: I think we're underestimating folks' ability to remember here! If the product actually is good, people will remember it.
- On being a real pain point: It's true that people have many "good enough" solutions to travel planning, but travel planning tools have failed because the past products really just haven’t been better by enough than using Google Sheets -- they assume that people don't like planning, or that they plan a certain way, and try to save time and automate it. In fact, people _love_ planning and sharing their trips! The tool just needs to work around the traveler, rather than dictate
Still, it's instructive that none of the links in the 2014 post's comments work. Will we prove them wrong? We can check back in 3 years when this article comes back to the front page again xD
You don't have to be just better than google sheets; you have to be a lot better than google sheets while also not introducing significantly more friction. Ideally none.
I think the space is hard because it involves tying together a number of different functionalities many of which are rarely done well, some which have never been done well.
Kudos to you for pursuing something that a lot of people are saying won't work. That takes some character! Especially in a difficult time for travel.
Your app looks great. As a full time traveler / nomad for 10 years, I don't have much of a need for itinerary planning apps on their own. But I would love to have social itineraries, where I can follow my friends' itineraries and share mine with them especially for serendipitously meeting up at random points around the world when our itineraries intersect. Yes, I've got an abnormally large number of globe trotting friends, but there are entire nomadic communities out there who are eager for this feature.
Are social features something you've considered developing further?
Hi! I downloaded your app, and convinced a friend who travels quite a bit to download it, on the basis of this post. It’s pretty great!
We have one piece of feedback: the trip page is very spread out, so there is no good way to see a quick overview of my entire trip, i.e. in city A from days x-y, then city B, etc. Unfortunately, a single day takes up the whole screen. If there was some way to have a more collapsed trip view in the app, that would be awesome.
This might as well be called the Age of the Blog Post, which seems obvious today in our age of Substack, Medium, and Wordpress, but back in 2004, definitely wasn't quite as clear!
> The Internet is changing that. Anyone can publish an essay on the Web, and it gets judged, as any writing should, by what it says, not who wrote it. Who are you to write about x? You are whatever you wrote.
That being said, I think the above has contributed a bunch to sensationalism. Unbundling of writing from magazines and newspapers into little articles incentivizes writing individual articles that stand out -- clickbait, shockers, and half-truths.
Good writing and crisp thinking definitely still rises to the top, but so does controversy and stirring up emotions. So yes, essays have won, but it's less clear the kind of essay that Paul mentions ("a few topics you've thought about a lot, and some ability to ferret out the unexpected") are the ones that have won
It's not even just video games! Leaflet's great for real-world non-video games maps of the whole world. It's by far the easiest system to make any random old image zoomable and pinchable.
My brother and I created a map of our local Costco in Figma and just put it up at https://costcomap.com/
One reason that often gets glossed over (including in this article) is how (non-online) Masters programs are huge moneymakers as a ticket to working in the US.
I've seen that in CS programs at University of Buffalo and Yale, a majority of Masters students are from abroad, especially from Asia.
Basically, if you wanted to work in tech in the US, you could either:
- Find a multinational company in your home country, work for them for a year, and request a transfer (highly competitive)
- Find a job for a company in the US while abroad, ask them to sponsor you for a H-1B, and hope you win the visa lottery (unlikely, since companies don't like their chances on the lottery)
- Get a Masters degree. Sure, it's $50,000, but you're guaranteed a 3-year work visa (OPT with STEM extension) during which you can apply for the H-1B multiple times (it's a winner!)
This is actually completely rational, and benefits both the students, the school, and arguably the US since it gets some highly skilled workers who did not take out loans.
Are Masters degrees outside of STEM (and online ones, like the OPM-managed ones this article mentions) that cater to local Americans in low wage sectors a scam? Maybe!
Are Masters degrees in STEM (which definitely are cash cows) a scam? Almost definitely not; in fact, they're a great workaround until the US reforms its immigration system
I would characterize working over talented foreigners in this way as a grift, rather than a scam.
They pay the high sticker price because it's rationally their best choice, but the value they're purchasing isn't in the degree, but rather the favorable status they earn in the immigration labyrinth.
This isn't good for them, except in comparison to other options. It certainly isn't good for native-born Americans, who are (for the most part) stuck paying the same high sticker prices, without getting the features which justify the high cost of the product.
It should also be viewed as potentially disadvantaging people already in the US. If these programs are catering to wealthy foreigners looking to immigrate, then we're not serving whatever need may be inside the country already.
In some fields, there is no need for masters degrees at all. For example in physics you can get a bachelors and career out, or get a PhD and be a physicist (industrial or academic), but there are no jobs I can think of that a masters would qualify you for that a bachelors wouldn't.
> but there are no jobs I can think of that a masters would qualify you for that a bachelors wouldn't.
Typically, community college faculty (and sometimes university lecturers) require a minimum of a master's in the field, and at least the first is as true of physics as it is generally.
In government it makes a difference in promotions and salary bands.
I have seen in electrical engineering some jobs advertise that they want someone with a masters, but really what they want is someone who understood and can use at least something from their undergrad and can be relied upon to exercise theory as opposed to someone who just passed.
I'm considering a masters in physics because I'm interested in possibly pursuing a Ph.D. there, but never got beyond 200-level in undergrad. The local university offers a physics MS that's basically a rebranding of their undergrad core, but at night to cater to working adults. It'd be cheaper to just take the undergrad classes, but that just doesn't fit into life, so I'm willing to pay for the privilege of being able to take them at night. (A Ph.D. program may never fit into real life either, but that's a different problem for a different day).
Some employers pay for higher education, but there is little ability for people to actually get that education. Many people in this thread do not value a masters specifically because they are not a good investment just looking at the price.
If the education system was different, more people would want them, and they would be more useful. See commenter below who notes that CS masters are usually quite useful but also are not employment gating and have many reasons to stay competitive with other programs.
Assuming you're actually smart and skilled - wouldn't it be better to just start your own consulting business, or if you're really ambitious, eventually an agency?
Aren't H1-B visa workers mostly in extremely HCOL areas and significantly underpaid / taken advantage of?
I imagine if you're from India and you really want to get out of India - this sounds like a good deal (although wouldn't $50k for tuition be hard to get?) But, I'm assuming most people just want to have more money / a better life?
Wouldn't the first option be better? And then you could stay closer to friends and family.
Consulting straight out of school is a really tough sell. Consultants are most often considered valuable by their employers because of their experience in a given industry.
I would not want to try to pitch myself as a consultant as a foreigner with little to no American work experience. If you’re doing it yourself, sales will be a challenge.
Heck, I did consulting in school to support tuition.
I think there are a few factors to consider.
(1) Ability isn't one-dimensional. Consulting is a different skillset than full-time work. It's not harder or easier -- you just have to be more of a package deal -- finding business, contracts, billing, etc.
(2) Interest. Most people are happier in a full time job not because consulting is hard but because they'd rather not do all those things. If you don't want to deal with contracts and billing, consulting isn't for you.
(3) Consulting goes all the way from cheap Indian contractors, through boutique UX shops, through Caltech professors consulting on esoteric issues.
You definitely don't need to be a rock star. Companies hire consultants for a variety of reasons. Some need some particularly hard technical problem, but plenty simply need a talent for a short amount of time. It's the same difference between in-house counsel (which is good for ongoing legal issues) versus law firm (which is good for one-off issues). Or if you're worried about a tree falling down on your house, you might bring in a random arborist to consult.
I picked up the skillset because I needed to pay tuition, not out of any innate interest or talent. You might not have the skillset for it, but it's not rocket science. You could pick it up too pretty quickly if the situation arose.
Now double the age, I find I don't actually like consulting very much. I need to charge about double to triple my salary to cover overhead, and overhead is all stuff I don't actually enjoy doing. It does have upsides. I do like the variety. The boom/bust of consulting works well if you have a bit saved up and a low cost-of-living -- booms are intense, and busts make for nice vacations. I also prefer not having to worry about things like overarching employment agreements and an employer owning me. But that doesn't outweigh the hassle of having to sell myself, send invoices, negotiate contracts, etc.
But not necessarily a technical rock star… If you’re really good at sales and you can somehow procure the technical talent elsewhere, you might be able to pull it off.
I am no rock star but my first three gigs after my MS were consulting gigs. 1 software and 2 network consulting jobs, including a high speed metro area network.
It is fun and flexible but not as nice as a steady job. Plus, I have no people skills. :)
"Smart and skilled" are only loosely correlated to success as consultant and small business.
This is not meant to be sarcasm. There are many other characteristics necessary and better indicators. I am smart skilled and successful. But I made a lousy consultant when I tried it too early in life - (self)salesmanship, business skills, tenacity and courage, people skills, and a specific view on risk acceptance are far more important. Relationships and a full Rolodex and branding / reputation don't hurt either.
(A specific view on integrity and ethics too. I'm not saying no integrity or no ethics. But the salespersons / deal closers in my area are all honourable truthful people whose job I couldn't do because my view of truth wouldn't necessarily correspond to theirs)
Finally, if we grant your last statement that most people want renumeration and happy life, for many smart and skilled people, stress and risk that comes with consultancy business does not contribute to happiness.
I used to work for a small tech company in the suburbs outside of St Louis, which is definitely not a HCOL area. When I left (to move back to a HCOL tech hub), probably about half of the engineering staff were H1-B visa holders. This was because we often had trouble finding qualified candidates that were either already local or willing to relocate to the area. Visa sponsorship was a much more compelling reason to relocate than anything we could offer US citizens.
The office was near Mastercard's global operations headquarters and it seemed like they also employed a lot of visa holders. So I don't think we were particularly unique in our willingness to offer visa sponsorship in a LCOL area.
>This was because we often had trouble finding qualified candidates that were either already local or willing to relocate to the area. Visa sponsorship was a much more compelling reason to relocate than anything we could offer US citizens.
Seems rather unlikely. What I guess you might mean is that you couldn't find qualified candidates at the pay ranges you were offering.
No, speaking from the Midwest there is in some areas a real shortage of experienced/qualified candidates. Even if you pay $200k, $300k, more, there are a lot of Americans who will not move to the Midwest. Folks from overseas, though, don't have the same preferences or prejudices. If you're coming from China, India, Bulgaria, Nigeria, Colombia, what does Minnesota vs Missouri vs Maryland really matter if you're coming for a job? Iowa vs Ohio? As long as you can find a suburb with good schools it's interchangeable. People who are attached to San Francisco or Seattle or the Northeast though do not harbor the same openness to moving to Minnesota, Iowa, Missouri, etc.
I'd love to see any midwest company paying those numbers, for a non C-level role. Hell, even trying to get above 100k/annum in the midwest is nearly impossible. Remote work options have made this a bit better; if the company then doesn't offer you "local market rates" bullshit.
Having driven through St. Louis (and Missouri) quite a number of times, I would be not be willing to relocate to the area. Maybe if the job paid really well AND was a really close match to what I wanted to do I might consider it. But for just pay, it's not even on my list of places I'm even looking for jobs.
That H1-B workers are "taken advantage of" is a myth. Yes some of them are but it isn't the case that if you hold such a visa you are being underpaid. Source: was an H1-B holder.
1. I think you are mixing smart with courageous, starting a business requires some courage.
2. Nope. At least those working for FANGs aren’t. You can see so many cases of folks were for low paying consultancies in India getting masters and then getting a job at FANG.
3. Getting an education loan for STEM is really easy and people are willing to put their parents house up for collateral.
I think with a FANG or good tech job you can be done with your loan in an year or two. Some people go to state schools and don’t get into any debt, or pay it off via internships. The US in most Indian people’s head comes with a better quality of life, a lot more money and better infrastructure. Perhaps some status too.
Right - I think a lot of people hope that by being more educated they'll get "smarter". That's not really how it works. You just get more educated.
In software, you DO need a lot of skills. But I think more important than that is having the ability to learn new things/skills quickly (being smart). The industry is always changing and evolving.
College is only going to get you the skills - maybe, a lot of times they teach you things that aren't that useful in the job market and don't teach you the things that are.
No, they're also abused in MCOL and LCOL areas. And they might be smart and skilled, but most of the ones I interviewed are lousy software engineers.
If we were talking about people with master's degrees from MIT or Stanford, it'd be a different ball game.
But I can't tell you how many candidates I've seen with something like a "Master of Science in Information Systems from the University of South Central Appalachia".
Well, even then framing a $50k degree as a workaround for US immigration woes, and not as an $50k worth of added value in terms of skills and education makes it sound like a racket, even if the numbers work out in the students' favour.
The gist is right, but details are wrong. Undergraduates can also get a 3-year work visa via OPT with STEM extension during which you can enter into the H-1B lottery multiple times (as many as four times if you time it well). It's just masters have higher chances in that lottery.
> are huge moneymakers as a ticket to working in the US.
Beyond that, the international Masters students are really subsidizing the education of other students. Every time there is some friction added to the student visa process, all but the top schools with huge endowment funds feel the financial pressure.
Most narratives I read about this situation paints the picture of the students buying their way into US residency via Masters program. But it works the other way too - many US citizens are able to get educated at an affordable price because of the money these Masters students bring in to the school.
we've had foreign job candidates with masters degrees in CS from top 50 schools who couldn't pass a simple programming test or who couldn't answer some basic CS questions on O(n) or whatnot.
I'm not talking sadistic google interview questions, these were things that anyone with a BS should be able to answer.
That has happened to me so many times that my last company shreds resumes from foreign students with a prestigious American masters where the undergrad is not similarly prestigious (eg IIT and Tsinghua are good, but not many other schools).
India has a population of well over a billion people. IITs admit only about 5000 students per academic year, from about 100K students applying, across all disciplines. If you find all other students from India not qualified, maybe there is something seriously wrong with your candidate outreach, or you hire from a very exclusive pool even stateside.
That's true, but having seen (non-IIT) Indian colleges, I am not surprised if 80%+ students they see are like that. Moreover, intelligent ones less commonly opt for masters degrees because they get good offers in campus placements, especially those from middle class or rural areas are already doing undergrad on loans.
We did only hire from about 20 schools in the US. The unfortunate thing is that the masters programs did not filter nearly as exclusively as the undergrad programs.
for "top 50 schools", what "top 50" are you implying? top 50 worldwide? top 50 us? or top 50 in their own foreign country?
As a sidenote: a lot of CS/Software Engineering graduates with a bachelor in my own country (Italy) can't pass simple programming tests, because most of the education is (too?) focused on theoretical aspects, and most of the time people is tested with non-practical exams.
I came to post the same thing - sometimes Americans forget that they are a country of immigrants. And one of the routes for potential immigrants is college education.
But with a typically enterprising capitalist mindset, the American system also wants to ensure that talented individual don't go back to their country (or elsewhere) immediately after getting a degree. This is where the high cost of a US college degree comes into the picture - the burden of paying for it acts like an anchor for most students who come from developing countries or econonomically weak background. Even if they get some kind of scholarship and / or do part-time work, they still have to take a huge loans to live in the US to complete their education. And often the fastest way to repay these loans is to work in the US or other developed nations. This may take another few years. The American system hopes that by then the potential immigrant would be sufficiently exposed to the American culture and lifestyle and consider staying here.
(The high cost also ensures quality of education is high in the US, thus attracting talents from around the world. And the money is also pumped into a lot of R&D in the college allowing US to maintain a big tech lead. It's a neat system that seems to work so far.)
The other aspect of ensuring that higher education remains costly in the US is to also ensure that a blue-collar workforce continues to exist, and wage is suppressed among the white-collars. Perhaps the law makers also feel that it acts like an incentive to work more diligently, out of anxiety and worry - after all, people with more qualification, experience and higher stable income often tend to jump around more (which the big tech try to thwart by entering into illegal agreements to not hire each others employee).
For what it's worth, when I joined Stripe (a bit closer to 400), I told my manager that I had a 2-week vacation after my first week of work, and they were totally fine with it! I suspect this is more of a "setting expectations early" and the cost of making a change
It's ridiculous that the only original source of this data, the IATA [0], charges $700+ for this list, so kudos to OpenFlights.
I can't stress just how important (and how hard) it is to get a great source of data for airports -- I've now built 3 travel-related projects (the latest, Wanderlog [https://wanderlog.com], keeps people's flight reservations, so uses it for an autocomplete), and it's been a key building block for all of them.
The main datasets we use are:
- OpenFlights [1]: mentioned in this post, but this dataset was great since it had timezone too.
- OurAirports [2]: no timezone here, but the "type" and "scheduled_service" columns in this dataset are essential. "Type" lets you distinguish between small/medium/large airports, and "scheduled_service" lets you easily filter out airports without real flights (which you often might not care about).
- Random other GitHub Gist [3]: I have no idea where this data comes from, but it was surprisingly complete and has a few golden nuggets like "num_flights" and "runway_length" in addition to "timezone". The presence of a "woeid" suggests Yahoo-related origins, but it's hard to be sure.
- We now supplement this with airports from autocomplete APIs like Skyscanner's, because they're still the most up-to-date.
Long story short, it'd be AWESOME to have one complete, updated database with all this data in one place. This kind of data really should be public and a public service, but until then it's unfortunately up to the community.
I've done something similar for my current job. I've used all these same data sources, even got access to the IATA stuff eventually. I also used GeoNames a lot, it's not specific to airports but it has decent airport data and I need a lot of the surrounding features as well.
Every source was definitely useful, but I think ultimately crawling Wikipedia was the most useful and highest quality set of data for me (after some significant data cleaning). The List of Airports By IATA Code [0] is almost as comprehensive as the official list from IATA, and you can follow the links to crawl info about the airport and city served. Getting info about what city the airport is considered to "serve" is so useful, as most airports are technically not in the city people consider them to be the major airports of, and some "serve" multiple cities.
Of course the difficult part there is that Wikipedia data isn't really clean or standardized. The page HTML isn't standard, even things that look very standardized like the sidebar will have 30 variations when you crawl all the airport pages. There is WikiData, but I found it still wasn't simple to get the data from there, and it also didn't include most of the page content which I wanted. [1]
Nowadays we have direct relationships with the airlines/GDS/so on, and also a department of people to add and manage the data ourselves, because even the direct source gives you pretty poor quality data. The project was way more fun when I was wrangling data from a dozen places around the web :) Now it's more of an enterprise CRUD webapp with some fancy localization and GIS tooling.
FlightStats / Cirium have an API for airport data[0] that I’ve found to be mostly complete (sans a few rural Australian airports). It includes historical records for airports that are no longer active, such as Hong Kong’s Kai Tak airport that previously went by the HKG IATA code.
FlightAware have a similar API[1].
These aren’t free or open mind you, but are at least readily accessible for those that need/want it.
Are you just looking for airports, routes, and schedules? FlightAware provides that: https://flightaware.com/
Not sure what you get with the commercial services, but even the free services are pretty good. It's what we used in 1st CAV to track the redeployment of the last units to leave Iraq in 2011.
It's only hard to get because IATA doesn't easily provide it. IATA isn't "a" provider of the data, they are the data. It would be like if you had to purchase a list of the bus stops and schedule in your city from your transportation department.
Many (most?) standards bodies charge for their standards documents and data feeds. There are obviously costs associated with running IATA; I don't see why they should be obliged to provide their data for free, especially when the typical user of such data is likely to build a for-profit business on top of it.
That’s a good point. The US has a somewhat unique system there. I’ve been googling this and I can find any definitive reference. All I can turn up is that the application for an IATA code is supposed to be made by an an airline or CSR, not the airport itself. What I can’t find is an explanation for the “coincidence” of the K prefix.
I have to pay the bus company to ride the bus, it doesn't seem insane that the bus company may want to charge me money if I asked them for a full, comprehensive, organized list of stops and schedules.
Sure, there are reasons for why they would want to make it available for free, but there are also reasons for why they would want to charge me, and they aren't unreasonable. I don't have any fundamental, natural right to a transportation network curating and providing their data for my consumption. It might not care enough about my needs to spend money do so. It might not care enough about my needs to spend money to do so for free.
This. From my YC experience, I know that starting with a niche is good advice and advice they likely would have gotten.
Looking at the Launch HN [1], I think the question should instead be "has YC lost it's way when its latest company is founded by 4 University of Waterloo students?". When phrased that way, the answer seems obvious, given that YC started out funding current college students in Boston.
Having gone through YC in W19 with Wanderlog (https://wanderlog.com), I think one of its best parts is how it invests in founders who are outside the typical VC hobnob. I remember meeting folks who'd worked in utilities in Texas all their lives, multiple groups of university students taking a leave, and other groups to whom VCs would simply say "you need more traction"
Like most early stage investors, they're not investing in an idea -- they're investing in founders! I do have to say that YC partners probably encouraged the founders to pivot (like we had been told), but the ultimate choice of what to pivot to is the founders'. So like another comment said, cut these 20-year old university students some slack!
Wanderlog (YC W19) | San Francisco, REMOTE (North America), VISA | Full-time Full-Stack Engineer | https://wanderlog.com/
Wanderlog (https://wanderlog.com) is building tools to make planning leisure travel easier. We believe that travel makes us and the world better, and are trying to lower the bar to it. We're one of the top-ranked travel planners on web and Android (and rising on iOS), and have grown even during COVID.
We’re currently a team of 6, including 4 engineers (counting twin-brother founders Peter, ex-Stripe; and Harry, ex-Googler as engineers), a designer, and a growth marketer. Our stack is Typescript with Node.js, React, React Native, and excellent front-end and back-end test coverage. We're looking for either an intern who can stay for 6+ months and a full-time engineer.
We also love traveling. Whether it’s a short hop to Austin, Seattle, or New Orleans; or a longer jaunt to Australia, Hawaii, or Banff National Park (all places members of the team have traveled to in the past year!), travel broadens our horizons, builds empathy, allows us to bond with others on our trip! We’re working to bring these experiences to more of the world.
I'm totally on board with the overall message that building (i.e. engineering) your own internal tools come with lots of overhead, but take issue with this one point:
> My counter-argument to that is there is also lock-in with internal systems. The most common version of this is the keeper of the spreadsheet.
The author then disparages spreadsheets as becoming the exclusive domain of one employee who wouldn't want processes to change.
In reality and my experience, though, spreadsheets are one of the most versatile and accessible systems, and their close cousins (Airtable, Notion) great as well! You can customize it to your own processes and they're pretty universally understood, so the barrier to change is pretty low.
Author here I actually think spreadsheets are great because they are so accessible. What isn’t great though is when there is a business processs that is a spreadsheet and knowledge that exists in one persons head.
The issue with excel is that there's no standard construction that anyone can query; most likely it's extremely hacked together since it's not a database, but made to look like one. Spreadsheets are great for doing stuff related to exports or before importing, but so, so often they lead to data islands built to either actively or passively protect someone's job.
> their close cousins (Airtable, Notion) great as well!
Their close cousins are locked, proprietary, slow, bloated, exceedingly complex, poorly designed (ui/ux), emojized, vc-backed feature extravaganza and have a subscription fee.
Experts of Excel use keyboard exclusively, their keystrokes are a melody of efficiency, expressivity and productivity that is continued to be mocked in similar fashion as the 2007-era Mac vs. PC advertisements.
I found Notion to be fairly simple to get started. Whereas when I opened Airtable my head exploded and I closed it immediately and haven't revisited it since.
I agree. Notion is a lot easier to get started on. I also should mention that Notion/Airtable vs. Excel are somewhat different types of swissarmy knives to be used in different environments. One is used in the battlefield and space exploration, the other one in a children's playground. Jokes aside, Notion has some nice features like Wiki and it wears too many hats like a joker in a circus. I love the fact that I can write a recipe with tables and markdown -> publish it to the internet to be consumed. Can't do that with Excel.
And it's also important to keep in mind that if your team is using various external tools, there is a high chance that only a couple of people will actually know how those tools work and even less of them will be able or happy to use those external tools.
This originally was big on Hacker News in 2014: https://news.ycombinator.com/item?id=8419658
This is really just the latest in a long list of articles about why travel planning startups aren't worth pursuing:
- https://www.phocuswire.com/Why-you-should-never-consider-a-t...
- https://paansm.medium.com/the-top-5-reasons-your-travel-star...
But we're still at it and have grown a ton over the past year!
- On frequency: I think we're underestimating folks' ability to remember here! If the product actually is good, people will remember it.
- On being a real pain point: It's true that people have many "good enough" solutions to travel planning, but travel planning tools have failed because the past products really just haven’t been better by enough than using Google Sheets -- they assume that people don't like planning, or that they plan a certain way, and try to save time and automate it. In fact, people _love_ planning and sharing their trips! The tool just needs to work around the traveler, rather than dictate
Still, it's instructive that none of the links in the 2014 post's comments work. Will we prove them wrong? We can check back in 3 years when this article comes back to the front page again xD