phsource

Wanderlog (YC W19) | San Francisco, CA | Full-stack Software Engineer, Front-end Software Engineer | Full-time

Wanderlog (https://wanderlog.com) is building tools to make leisure travel easier. We believe that travel makes us and the world better, and are trying to lower the bar to it. Our product so far is a Google Docs for planning travel and sharing recommendations.

We’re currently a team of 5, including 4 engineers (counting twin-brother founders Peter and Harry as engineers) and 1 designer. Our stack is Typescript with Node.js, React, React Native, and excellent front-end and back-end test coverage.

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.

If you're interested in joining us, email me at peter@wanderlog.com or check out our job posting at https://wanderlog.com/blog/jobs/full-stack-engineer/

I'm a bit confused: the featured link here seems to be made by Google, and seems pretty performant in Firefox on a 6-year-old Macbook (albeit with a good Internet connection). Were you thinking of other links that had been shared before?

> Copy [means] to borrow an idea for its known useful results; steal [means] to take ownership of an idea and extend it to create some novel result.

> Copy the same sponge any cake could use, then using your own creative talents as icing to make something uniquely great (steal) in the areas that make a difference in your product's niche... Steal strategy! Copy tactics

As an engineer working on product, this is good advice. It's often appealing to reinvent the wheel and succumb to "not-built-here" symptom. Especially when it comes to questions like "what stack and language to use", often, the best answer is "whatever has worked before." Rather, what's worth reinventing is the actual features or user-facing aspects of the product.

Don't go crazy making your own global Javascript store: just roll Redux or MobX. As someone who loves engineering challenges, I find myself reining myself in a lot on this at Wanderlog (https://wanderlog.com). But I realize whether we're successful or not isn't going to be our technical infrastructure. That needs be solid, but what really matters are the little things we can do to wow travelers

> In particular, it is often difficult to make the (important) decisions about the organisation of the argument, and selection of good notation, until a large part of the paper is already written

This is very true. When writing a blog post in China during the earlier parts of coronavirus [1], I had drafted an initial version where I just went through my experiences in chronological order. Only after writing it all down did themes emerge in how I finally structured the article.

This really speaks to the power of interdisciplinary concepts though. Writing an essay by writing an outline first and then filling in the details probably predates rapid prototyping by eons. It's cool that it all circles back to writing academic papers only through this leap!

[1] https://wanderlog.com/blog/2020/01/29/coronavirus-observatio...

The advice we get from Y Combinator (which I think is extremely valid) is to focus on our product and customers, because that's ultimately what investors are going to look at. Fundraising is distracting, emotionally draining, and unproductive if you don't have those. If you're spending half your time during the program fundraising, that's half the time not spent on more important tasks

Suhail from Mixpanel (S19) and Laks from Zenefits (W19, my batch) were both in recent batches! So they definitely exist. Much more common is the non-huge-exit but still successful second-time founders

I'm glad they're moving Demo Day up. The whole point of YC is to level the playing field for all kinds of founders, and this would help keep the focus on Demo Day rather than the behind-the-scene investor meetings that tend to benefit connected, established founders.

Personally, as a YC W19, first-time founder of Wanderlog [1], I didn't do any fundraising before Demo Day. But I heard from friends in the batch about cold emails, "informal" dinners, and all kinds of shenanigans to get ahold of founders earlier.

Demo Day helps even things out by giving attention to all startups in YC, and not just the companies with connections. Everyone's given the same amount of time on stage, and off stage (in any other year), everybody's milling about in a huge hall. You can't tell who's the recent college dropout, and who's a former co-founder of a co. with a billion-dollar exit.

I can totally see investors really taking the coronavirus/lack of an in-person Demo Day [2] as an excuse to double-up on their outreach, leading to an even more imbalanced process. Good work on the YC Partners' part for short-circuiting this and helping the first-time founders and the traditionally disadvantaged

[1] A trip planner for vacation itineraries: https://wanderlog.com

[2] https://blog.ycombinator.com/yc-w20-online-demo-day-now-on-m...

Looking through the Wayback Machine, it seems like this was a funky styled-email service that allows sending emails with backgrounds, emojis, etc.:

http://web.archive.org/web/20200225122954/http://www.incredi...

Definitely something whose time seems to have passed, and probably legacy for a couple of years judging by the "Windows 7" support badge on the bottom; I'm not quite sure how this is newsworthy, but then again, it's always interesting to see a relic from another age

I am curious if there are other YC founders in the crowd -- did you find the in-person part of Demo Day very helpful, or did more happen online before/after?

Sorry about that! We just fixed it -- if you create it again, hopefully you get a non-Vietnam War picture

The interesting thing is the "Agony" sort has almost become standard! Skyscanner, Momondo, and others all default to a "Best" sort that takes into account flight length and prices.

Google Flights also does this for the first few results it returns, before returning a bunch of flights in price sort order.

This is a long time coming: for more context, Hipmunk was acquired by Concur back in 2016 for $58MM [1] as the flights meta-search business was just getting tougher and tougher

As someone who's dabbled in flights bookings (BookWithMatrix) and now am working on a travel startup (Wanderlog, https://wanderlog.com, a Google Docs/Trips for travel planning), flights are just not that profitable. In North America, you'd be lucky to get 1-2% commissions. In the rest of the world, it's a bit better, but Hipmunk was definitely North America-first.

Concur never was a consumer company and Hipmunk never was going to make a ton for them, so it was just a matter of time. RIP -- a lot of their best features (not just showing the cheapest flights, but the "Best" based on agony) have made it into all major meta-search tools, so it's not for nothing!

(Also, I personally use a combination of ITA Matrix, Google Flights, and Skyscanner now. ITA Matrix still has time bars, if you loved that UI; Google Flights is just so fast, and Skyscanner searches low-cost carriers that Google Flights sometimes misses)

[1] https://www.crunchbase.com/organization/hipmunk

Ah that's unfortunate; we did find it a bit weird that you have to think of all changes as `ops`, but otherwise, I'm curious as to what you've had issues with.

We also have a custom back-end that's MySQL that may have changed how it work a bit

Wanderlog (https://wanderlog.com) is a Google Docs for planning travel, and naturally, we had to figure this out early in the process.

If you're on a Node.js/React stack, we highly recommend using the combination of OT-JSON0 [1] and ShareDB [2], two excellent libraries. OT-JSON0 lets you perform operational transforms on any JSON-serializable structure pretty intuitively, and ShareDB handles synchronizing it between multiple clients and resolving conflicts. Building your own is going to be a pain, so unless your core business is a text editor like TinyMCE, it's probably best to stand on the shoulders of giants.

Another approach that we didn't end up using is Figma's: especially if you're working on anything other than text editing, ask if you actually want to use operational transform. Something simpler might be good enough -- in Figma's case, they use "last write wins" because they were able to break the document into small enough, atomic elements. [3]

[1] https://www.npmjs.com/package/ot-json0

[2] https://www.npmjs.com/package/sharedb

[3] https://www.figma.com/blog/how-figmas-multiplayer-technology...

Have some empathy! I've seen a trend where rants against the modern web often get upvoted lots on Hacker News. Consider this quote from the article:

> I make no secret of hating the mouse. I think it's a crime. I think it's stifling humanitys progress, a gimmick we can't get over.

Does the world's typical computer user today hate the mouse, and prefer a keyboard-only interface (CLI)? No -- in fact, command-line interfaces are less discoverable and harder to use, starting out. Even as a programmer, I struggle to remember the flags to many common command line utilities

Sure, the author's example of a cashier's checkout console might be great as a text-only interface -- cashiers use it day-in, day-out, and can learn all the keyboard shortcuts in a day. But what about the self-checkout machines that shoppers use maybe once a week? Would you rather have every person have to learn a list of keyboard commands while navigating a two-color interface?

Does the modern web poorly serve the author, who's good enough with technology to master any UI? Sure!

But the modern web works better for the billions who otherwise would not have started using it in the first place

Plaid, Yodlee, and others abstracts away extracting data from various banks and financial services providers, so they're providing a solution built on top of the same data extraction techniques that this tool uses

This is very cool! I love how you brought back the original Kimono UI with the checkmark and Xs for adding and removing data tags.

We built WrapAPI (https://wrapapi.com) back in the day, before we ended up starting Wanderlog (https://wanderlog.com), our current travel planning Y Combinator startup. This definitely is still an unsolved problem.

However, from a business point of view, we found that it was rather difficult to make a business out of an unspecialized scraping tool. The Kimono founders expressed a similar sentiment: ultimately, scraping is a solution looking for a problem.

Developers can often roll their own solution too, which limits your customer base and how much you can charge. Instead, vertical-specific tools that target particular industries seem to be the way to go (see Plaid as an example!)

Alternatively, you have to be good at Enterprise and B2B sales. This is a product that you need to get the word out, get a champion, and do customer success on since it has a substantial learning curve. We were not, so that was why we chose to focus on other projects to start out

Best of luck, and feel free to get in touch if you'd like to chat more

I have to say, being able to add all the places to a list accessible from Google Maps is really key since it's one of my main tools while on a trip. I'm curious though -- what got you to choose exporting to Google My Maps vs. a Google Maps saved places place list?

For Wanderlog (https://wanderlog.com), we ended up exporting to Saved Places instead of Google My Maps. My Maps lets you do advanced things like create regions and polygons, but when using it on the Google Maps app, it seems strictly worse:

1. You have to explicitly turn on the My Maps layer, rather than the pins always showing, for it to show up on the app, and

2. When you tap on one of the My Maps pins, it doesn't show the underlying place's associated details (e.g., rating, etc.) which makes it a bit harder to go back to the source of the info.

It's definitely nicer to build on top of Saved Places, which also seems more actively maintained by Google. Did you have any reason in mind to use My Maps rather than the Saved Places?

I didn't realize the significance of this, but searched a bit more, and found this:

> OpenBSD/loongson is a port intended to run on systems based upon the Loongson (also known as Godson) 2E and 2F MIPS-compatible processors, using the PMON firmware and boot loader

> ...

> Supported hardware:

> Lemote Yeeloong netbook [1]

The Lemote Yeeloong rang a bell, since it's one of the few computers that's "built on free software from the BIOS up" that was recommended by Richard Stallman [2]. Glad the effort to run this is still going strong and supporting the latest version of OpenBSD, released in May!

[1] https://www.openbsd.org/loongson.html

[2] https://en.wikipedia.org/wiki/Lemote

Ah, that makes a lot of sense! I missed that part of the article. It sounds like the decision is safe in Massachusetts, then!

While I personally agree with the ruling, there's a good chance that it will get appealed (and potentially overturned), since US v. Carpenter, decided back in 2017, had a narrow holding.

The article mentions that the EFF:

> ...asked the court to recognize, as the Supreme Court did in U.S. v Carpenter, that people have a constitutional right to privacy in their physical movements.

But SCOTUSblog's reporting of that decision makes it clear that [1]:

> Roberts also left open the possibility that law-enforcement officials might not need a warrant to obtain cell-site location records for a shorter period of time than the seven days at issue in Carpenter’s case – which might allow them to get information about where someone was on the day of a crime, for example.

There are two dimensions involved: who's being tracked, and how long it's been tracked. In Carpenter's case, someone was tracked for 7 days without a warrant, while in Almonor, it was only 6 hours. Additionally, the EFF post suggests that this was a much more targeted search (one for specific suspects only) than the one in Carpenter.

As much as I dislike the idea of warrantless phone tracking, since Roberts crossed over to vote with the liberals on the side of a narrow majority on Carpenter, there's a decent chance this might get overturned on appeal.

[1] https://www.scotusblog.com/2018/06/opinion-analysis-court-ho...

> Significantly, they also acknowledged the contributions I made. Acknowledgment doesn’t require a parade. These are small gestures that folks made that were meaningful to me:

> Using the :clap: emoji in a pull request comment to highlight a clever bit of my code.

> Sending a direct message of thanks upon realizing they were actively benefiting from a refactor I’d made to a previously gnarly bit of code.

> Posting a note to our #gratitude channel on Slack recognizing how helpful some documentation I’d written had been to them, and encouraging others to use it.

I remember reading a study about how for a romantic relationship to succeed, the ratio of positive to negative interactions has to be 5 to 1. [1]

Work relationships might only need a ratio of 1:1, but especially as an engineer, where the default's to critique and only pick out the bad parts in code reviews, these really help! Even at Stripe, I remember how far a "Nice! I've been meaning to do this for a while" as a part of a code review would go in just making all the other suggestions go down better. Definitely kudos to CircleCI for making this a part of the culture.

[1] https://www.gottman.com/blog/the-magic-relationship-ratio-ac...

Thanks for trying it out in such detail!

> 2. Input validation needed for links

We just fixed this; thanks for the report!

> 3. Is it possible to scope searches?

I just added search scoping based on the current map viewport; thanks for the suggestion! I've been meaning to do this for a while.

> 4. Default map zoom is inconsistent.

I also just pushed a fix that zooms in if the newly-created marker is too close to another one. Take a look and see if it works!

The other issues will take a bit longer, but they're all definitely real:

1. We have issues with editors editing text at the same time, and this is an ongoing limitation of how we're using ShareDB on the back-end. We'll do our best to make it better

5. We definitely should add bubbles to show who else is editing for sure! You're not missing anything -- we really need to do this

On the overall business feedback part, it's definitely tough. We've had some success with affiliate ads for our other sites, but otherwise, B2B is a much easier market in many ways. We're going to build the best product we can, and keep costs low.

On the travel agents side, one name you've probably heard of is Travefy! (They similarly pivoted from a B2C to B2B) Take a look, and feel free to give me an email if you'd like to be introduced to a Travefy early-user travel agent to learn what she likes about it (peter@travelchime.com)

We have them (for some cities)! Check out https://travelchime.com/plan/peelkusfqahdnjxk/list/2987 (they're one of the lists that show up at the top of the page)

Site's down, so alternative (Google Cached) link: http://webcache.googleusercontent.com/search?q=cache:paAIFbe...

Definitely not, from experience. I think the opening poster means that Adyen needs such a large reserve, since I definitely know subscriptions-based startups that don't have any reserve with Stripe

It seems like Waymo's hitting a point of diminishing returns. The past 4 years' worth of disengagements per 1,000 miles driven were:

- 2014 (last 3 months): 1.27 [1]

- 2015: 0.80 [2]

- 2016: 0.20 [2]

- 2017: 0.18 [3]

There are two possible causes off the top of my mind:

1. Waymo's cars are being tested in more demanding environments.

2. There's a floor to the number of disengagements due to technology and road conditions: situations that are different/unique enough that no model could accurately detect/respond to it.

Case 2 would be definitely be a problem for the full-autonomous (i.e. zero user-input) type of car. I'd love to hear other hypotheses/feedback from those closer to this kind of work!

Also, another interesting note: total miles driven on public roads is actually down from 2016 (352k vs. 636k earlier). Perhaps Waymo's finding the simulated/private-road tests more fruitful for winnowing down the final edge cases.

[1] https://www.dmv.ca.gov/portal/wcm/connect/dff67186-70dd-4042...

[2] https://www.dmv.ca.gov/portal/wcm/connect/946b3502-c959-4e3b...

[3] https://www.dmv.ca.gov/portal/wcm/connect/42aff875-7ab1-4115...

I love how this story weaves the changes of an industry around a single life who anticipated, benefited, and eventually fell behind the trends of how we get music, movies, and media. I still remember buying bootleg "VCDs" in Hong Kong back in 2001.

When it comes to how the music industry's changed as a whole, I think two charts tell the story:

1. RIAA-tracked (North American) music sales have fallen from a high of $14 billion in 1999 to only about $6 billion today, even with downloads included. [1]

2. At the same time, concert ticket sales have gone up from $1.5 to $7.5 billion, offsetting three-quarters of the drop in music sales. [2]

It's an incredible shift in how the industry works, and explains the rise of Live Nation/Ticketmaster, Spotify (why are music publishers and others are more willing to give away music?). The winners and losers have changed, but the music keeps playing.

[1] https://blog.thecurrent.org/2014/02/40-years-of-album-sales-...

[2] https://www.statista.com/statistics/306065/concert-ticket-sa...

This seems to be a sour grapes article representing the grudges of online travel agencies (OTAs) against Google. This quote really caught my attention:

"OTAs earn a roughly 20% commission from hotels for each reservation they book, which covers their cost of marketing, inventory acquisition, customer support and payment processing."

20% is a ton! Later in the editorial, the writers complain about how Google takes "a 10% to 15% commission on net revenues from reservations booked through its meta-search. Google thus gobbles up most of the profits that OTAs earn".

But Google is taking on the marketing! To be fair, inventory acquisition and payments are also costs, but from running travel sites before, travel truly is a marketing game. As a result, their take makes sense given that they're really taking the place of the traditional marketing channel.

Google's concentration of power definitely is concerning. But the OTAs (who have been living off fat profits for years; see Priceline stock [1]) get none of my sympathy.

[1] https://finance.yahoo.com/quote/PCLN/; has outperformed Google stock over the past 10 years

Good catch here on the confirmation email link issue! I just fixed that. I've noticed for creating an email subscription, it takes a while for the overlay to close; did it never close for you? I wasn't able to reproduce this, so let me know if you have more details.