ezl

yes, fair enough. even without the 2 way obfuscation, I'd still pay for this.

craigslist implements something like this now. If you email a CL anonymized email address and email back and forth, you never get the counterparty's email address.

Basically craigslist just acts as a relay, so all communication actually happens between $PERSON_A and Craigslist and separately between $PERSON_B and Craigslist.

As far as I know, 10minutemail does not do this. I personally use 33mail for this sort of forwarding.

However these are designed around obscuring MY email address, and using 10minutemail.com or 33mail.com's domains.

I need to create emails for my customers that will forward to THEIR email addresses when another party contacts them, preferably using my own domain.

While I guess I technically could figure out a way to use something like 10minutemail and register accounts for all of the users of my site, it would probably require:

1. the users of my site to tell 10minutemail that it's ok to forward to them (during 10minutemail's registration process)

2. i would have to post 10minutemail.com domains as the contact info for the customers of my site, which is an undesirable UI.

If I'm missing something and I can in fact use existing services to do what I want, please fill me in, because I want this, I just don't want to build it (yet, anyways).

I will pay for Craigslist style email obfuscation as a service.

Problem:

I have a marketplace type website where my users would like to be contacted but not have their emails be made public.

How it works:

I set a subdomain of my site, mail.mysite.com to an ip address you designate. On demand, i make an API call to your service (www.emailobfuscationasaservice.com) with

    {
    real_address:foo@bar.com,
    expiration_timestamp:2014-01-20T00:00:00.000Z,
    obfuscate_replies: true / false,
    callback_url: (optional)
    }

www.emailobfuscationasaservice.com responds with either:

(A) a success message and an anonymized, temporary email address like 239fag72wa@mail.mysite.com (the subdomain i gave you), or

(B) an error message like "no such user at that domain" or "domain doesn't exist" or something.

The email address lives until the expiration timestamp. If the obfuscated email receives a message, you forward it to the "real_address".

If "obfuscate_replies" was set to true at email creation time, then if the person responds, you play middleman for both parties and force all their emails to go through you (basically you create another obscured email for person who writes to the obscured email).

If "callback_url" was set, that means that I, the operator of www.mysite.com, want to be notified of activity at the obscured email address, so you POST activity to the specified url. That lets me do things like update activity feeds in my app for the user.

If you build this, I will pay for it. It seems to me that many other startups could use something like this as well.

edit: formatting and owning up that this is me re-posting this from https://news.ycombinator.com/item?id=7024718 in hopes that somebody will build this so I can pay for it.

I'm curious about what one of these decks looks like.

When you raise money, you presumably have some notion of how you are going to spend that money. $X for new engineers, $Y to build out the London office, etc.

What does a plan for SPENDING the $80m they raised in this round even look like?

brokers often do not fulfill their fiduciary responsibility.

It's not uncommon at all and I've seen it happen quite openly in many situations. It's tricky business outside of the pure electronic game.

chris, you're right in the context of HFT stuff, but of course different matching engines have different rules.

There was a big trade for a while where people had different account types. the CBOT matching engine, for example, had customer orders prioritized over marketmaker or firm, so customers actually COULD jump the line.

Some companies had customer accounts specifically so they could insert orders in higher priority in the queue (the tradeoff is that customers pay for CXLs, but you can just do the math to see when you should be using which account)

to @panarky of the comment, its not on the order of microseconds -- the stuff people used to rip on were flash orders, which are exactly 30 milliseconds.

So @panarky has a bit of a point about knowing the order in advance, but its still damn hard to profit from it. The narrative of just front running trades by having order information in front is just not feasible because of the bid-ask spread (the flash orders are to try to help maintain some semblance of BBO consistency across exchanges)

So as yummyfajitas says, it's not possible to sell ahead (saying "short ahead" in that way exposes you as not really knowing the trade, btw) or buy ahead in a way that necessarily affects price.

Phil, I wrote a response but HN rejected it because it was too long, so I posted it on my blog.

http://blog.ezliu.com/which-problems-to-solve-first-in-real-...

TLDR:

Despite directly opposing headlines and section titles, I don't think we really disagree on much.

Real estate is a big space and there are easy and hard problems. Most of our opposing headlines are a result of us describing different aspects of the game.

Happy to chat any time. Email in profile.

@followsteph: a ton of respect for what you've done.

How do you acquire most of your customers?

What channels have been most effective for growth?

regulation can be a pain as you say -- there are a lot of highly regulated components of real estate, but some of them are pretty easy to get into as well. For example, crafting rental property management software like maintenance ticketing systems.

Some interesting companies in the residential home sales game are trying to streamline the process. Off the top of my head, paperless pipeline & dotloop are both trying to reduce the paper volume.

there was also a brandery.org (cincy incubator) company in the last cycle that was doing something similar.

Padmapper is really awesome. I think one of the reasons it were so successful was that it gave an alternate (better) UI for apartment search.

People are visual and they know maps, but seeing lists is harder to parse.

The fact that nobody else at that time had put apartment search on a map is a great example of how low the bar is for excellence in this space (no offense, @padmapper).

FWIW, I recommend www.rentshare.com for this.

I use them to pay and receive rent (I'm a landlord and I rent my residence).

Additionally, whenever landlords who use Rocket Lease (my rental application company) inquire about rent payment online, I typically recommend Rent Share.

They'll even accept electronic payment from the tenant and mail a paper check to the landlord on your behalf, so if you want to have it autodrafted for your rent, you don't even require landlord opt in.

To reinforce your comment:

Some of our customers have been bringing iPads to showings and asking prospective applicants to fill out digital applications on the spot.

To put a number on it:

6% transaction fees you're talking about are probably in home sales. About 800B a year transacts in the US in home sales (existing home sales data: http://www.realtor.org/sites/default/files/reports/2013/emba...). In practice, that 6% number often is a bit lower for the brokers. Also, FWIW, on top of broker fees, you'll be paying attorneys, government taxes, etc.

In Chicago, to transact in and out of a house, you should expect to pay between 10 and 11% (so if you buy a house for 100k and sell it for 110k, you will break even if you're lucky).

I completely agree.

Apartment search is, in my opinion, the hardest rental vertical to try to tackle, mostly because solutions the 2 sided market traction problem have been elusive.

It also seems to be the most common one that founders attempt.

In contrast, rent applications, management software, online rent payment portals are value providing from customer #1.

@ 1. Craigslist style email obfuscation as a service.

problem: i have a marketplace type website where my users would like to be contacted but not have their emails be made public.

how it works: i set a subdomain of my site, mail.mysite.com to an ip address you designate. on demand, i make an api call to your service (emailobfuscationasaservice.com) with

    {
    real_address:foo@bar.com,
    expiration_timestamp:2014-01-20T00:00:00.000Z,
    obfuscate_replies: true / false,
    callback_url: (optional)
    }

emailobfuscationasaservice.com responds with either a success message and an anonymized, temporary email address like 239fag72wa@mail.mysite.com (the subdomain i gave you) or an error message like "no such user at that domain" or "domain doesn't exist" or something.

the email address lives until the expiration timestamp. if the obfuscated email receives a message, you forward it to the "real_address".

if "obfuscate_replies" was set to true at email creation time, then if the person responds, you play middleman for both parties and force all their emails to go through you (basically you create another obscured email for person who writes to the obscured email).

if "callback_url" was set, that means that I, the operator of mysite.com, want to be notified of activity at the obscured email address, so you POST activity to the specified url. that lets me do things like update activity feeds in my app for the user.

If you build this, I will pay for it.

@ 2. support page as a service.

copy these: http://supportdetails.com/ http://www.browser-details.com/

Sell more aggressively.

@ 3. Basecamp for travel planning (I imagine this already exists).

I am on a trip with some friends right now. We have ideas for stuff we want to do but its all poorly organized in email and google spreadsheets. Basically the things we want to do are: tourist sites, activities, food.

A cool app would let us create a group like in basecamp, state our destination, then let us start typing things in we want to do. those get matched to existing review sites or links that contain information about them and put on a map.

in an unfamiliar place this is useful because it lets me group items/events geographically and plan my trip better. it would let people plan an itinerary together, asyncronously and make comments on possible activities for the group.

(other smart features: places to stay, modes of transportation, notes like "don't forget to get your visa by date XXX").

Basically a crud app. how to monetize? no idea. affiliate commissions on airbnb, event tickets, etc? i want it, wouldn't build it. the backend isn't hard, getting the UI right would be a challenge.

Love this comment.

Great UI is about words.

Reminded me of this: http://37signals.com/svn/posts/3404-reminder-design-is-still...

I normally love SA's writing, but this opinion piece is really narrow minded and I think misses the point.

cs702 had a fantastic comment a few days ago in another thread [1]:

    Bitcoin befuddles experts who analyze it from a narrow perspective,
    because  it is not just a new medium of exchange or a new store of
    value: it is also a new kind of point-of-sale payment system (one
    that doesn't require payment processors), a new kind of global
    financial transfer system (one that doesn't require financial
    institutions), a new kind of time-stamping certification system
    (one that doesn't require notaries or county clerks), a new kind of
    contract-enforcing mechanism (one that doesn't require lawyers), etc.

    With rising global adoption, many new kinds of applications are likely
    to be created to take advantage of the Bitcoin network, the design of
    which even specifies a built-in script for defining and executing new
    types of transactions involving any arbitrary number of parties.

    In short, Bitcoin is a technology platform -- one that is benefiting
    from network effects.

    It may fail as "money" (in a narrow sense) and still succeed as a
    global platform.

Even from SA's narrow perspective, it's not clear why "legitimate transactions" is a requirement for success. As long as there is a large enough community of people who want to use it to transfer value for any reason, it'll continue to sustain itself.

The one takeaway he offers in closing is spot on though: "Just as it’d be stupid to convert all your dollars to bitcoin, it’d be stupid to not pay attention."

[1] https://news.ycombinator.com/item?id=6810839

I've worked with Ryan several times now and referred him to friends who have all informed me afterwards that they intend to work with him again when they have an opportunity.

Ryan has:

1. a slow, thoughtful approach to UI/UX design for your product. he wants to genuinely understand your needs first

2. meticulous attention to detail (unreasonably so, which is good because it lets me be lazier)

3. writes reusable, clean, well commented code (javascript and well structured html and well organized sass),

4. makes working with him easy/low time cost by writing to your codebase and committing to your django/rails project directly in a fork, rather than forcing you to take html/css/js and adapt it back into your templating system

If you're looking for a front end guy who will think about your product from the perspective of another founder rather than a contractor trying to squeeze you for a week of pay, you need to at least start a conversation with him.

Happy to provide more details on my experience working with him as well, just drop me an email.

i found it interesting that depreciation is expected on your property and houses are expected to have no resale value.

Zac Brown Band - keep me in mind

Bid simulator is nice, but what I really wanted to highlight was the existence of profit maxima and the fact that it can be determined experimentally.

In retrospect, Bid simulator is probably the best thing to use for a starting estimate of the keyword profit optimum (if you don't have data and just use your global conversion rate across all keywords.)

Obviously google has the MOST information and is in the best position to optimize on advertisers behalf, but among SEMs I trust and respect, I've found that there tends to be a distrust for how Google applies the information and a belief that despite the inferior information, a more hands on approach tends to outperform Google's built in tools.

For example, Google doesn't use the keyword when determining what ad to show in its ad rotation optimization. http://searchengineland.com/optimize-for-conversions-using-t...

Google's 2009 internal research indicates this isn't true:

http://adwords.blogspot.com.au/2009/08/conversion-rates-dont...

"...on average, there is very little variation in conversion rates by position for the same ad. For example, for pages where 11 ads are shown the conversion rate varies by less than 5% across positions. In other words, an ad that had a 1.0% conversion rate in the best position, would have about a 0.95% conversion rate in the worst position, on average."

there are general techniques for "seeding" search terms.

Obviously the best approach is to have someone manually do it who understands the space to initially seed it. However, its hard to get a really nice long tail of keywords that way.

The way to do something like this in an automated fashion would be to just find the competitor products and use something like SEM Rush to deduce what keywords they're hitting, then use a keyword expansion tool to expand that list.

After the initial setup, you should always be spending some of your adspend on expanding your keyword list. for example: 80% of your budget goes towards "exploit", which tries to perform as optimally as possible, and 20% of your budget goes towards "explore", which is basically letting Google and users give you hints about what words might be profitable.

To do this, create another campaign with broad match keywords (google matches a bunch of crap to these). Your script should regularly take the queries that match the broad match terms, then either:

(1) negative match it so you never see it again, for bad keywords

(2) exact or phrase match it to be managed by the CPC bid optimization method described in the post.

http://www.wordstream.com/products for example does some of this (and google itself will give you keyword recommendations)

the price of bitcoin is skyrocketing.

why? "more buyers than sellers"

the square blog post I referenced implements this better. they save tuples of (timestamp, location) and can therefore determine velocity (more or less).

I used a really ghetto distance metric as a proxy for velocity and since the sampling rate varies on everyone's local machines it can produce really ugly signatures.

close inspection of the implementation will not bode well for anyone's opinion of my programming abilities.

my excuse is that this is a tiny bit of UI candy on top of my core business. if i were to spend much more time on it, I'd change a lot of things about it.

I did.

I really wanted to use it and have emailed back and forth with them several times over the last 2 years (from before they even had a product called hellosign -- when it was integrated into hellofax).

it required account creation for both signers through their site, which forces my users to leave the site to sign, and it wasn't a fit for the UX i wanted.

Similar issue for signnow.com

Benefits of using those though: transfering the responsibility of maintaining legally compliant signatures to a third party. Read more: http://www.hellosign.com/info/legal

For my specific use that's less important, so I opted for a solution I could embed without sending my users somewhere else.

However I think the hellosign folks are awesome guys and i use hellofax for a lot of stuff personally.

I did look at it. It was my second choice and I would have used that if I didn't used Thomas Bradley's.

I emailed Thomas Bradley when I was exploring options and he responded within an hour. Also, that repository was active (TB was responding to bug reports and there had been commits on the day that I looked) and had more followers and forks. Since I'm a mediocre hacker, I generally optimize for paths where I can get the most support if I think I might need it.

Thomas's is also very well documented with specific code examples and demos for the basic use cases.

I considered a lot of different options, including signature as a service (https://www.signature.io/ -- these guys have a beautiful implementation), integration with other companies (hellosign and signnow are my favorites, docusign/echosign), but TB's was just the best fit for my needs at the time.

I was an organic chemistry TA for a professor who was notorious for having the hardest organic chemistry class at our school, but it was also widely considered that surviving his course meant that you truly understood the subject matter.

He was unapologetic about the subject matter being difficult and believed that teachers who didn't teach it hard and try to really force undergrads to understand it were cheating themselves.

He was especially adamant about it because many undergrads taking organic are trying to go to medical school and didn't like the idea of letting lazy or stupid people become doctors on his watch.

When we were grading tests, he would laugh as he marked errors and delivered low grades. "Hexavalent carbon? WRONG. Saving lives... Saving lives..."

fwiw this really resonated for me because i'm a single founder in a semi-crowded space. my strategy for success is to basically have a much smaller scope and very narrow feature set, but to be crushingly better than the competition in everything that i DO attempt.

or maybe i'm just seeking out articles that make me feel good about myself.

For the deals I've offered, it was always sales - variable cost of sales.

I sell credit reports. So just take out the vendor cost of the reports.

I am happy to eat credit card fees, server fees, any other fixed costs for running the business, salaries etc. to test candidates if it helps me find good people.