Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Ask HN: I built it, nobody came, now what?

news.ycombinator.com

Post reply on HN
Print view View original

241–250 of 270 posts

Re: Ask HN: I built it, nobody came, now what?

#241

Post by d--b »

sjg007 wrote: post #238

Looks like a great tool. I know people who have built similar things in Java for example... You are competing with Jupiter notebooks and R in some sense here. Is your market business analyst? Data scientist or something else? How expressive is your DSL and how easy is it to learn? A few other things, can I export one of the models I develop to a web server so that I could deploy it to my internal network for someone…

yes business analysts, quant/devs, people with some coding experience who talk to business guys.

This DSL is javascript. It's not the easiest of languages tbh, but it's better than it used to be. If you're curious the runtime is based on www.observablehq.com, these guys run it as a notebook.

Right now, you can't export to a website, but that's the plan (for the paid version - still cheap, ha).

Export to Powerpoint / Pdf is definitely something I'll want to add in the not-too-distant future.

Connectivity to PostgreSQl, is something I am working on right now.

Re: Ask HN: I built it, nobody came, now what?

#242

Post by d--b »

chmod775 wrote: post #220

I had a quick look and for your benefit I'll be blunt - I hope you'll see it that way. Your application directly competes with Excel and is both more complicated and less powerful at the same time. The JS syntax is too much for your average MBA. To avoid competing with Excel, which thousands before you have attempted and failed at, you need a way to make it interoperable with Excel - that means an easy way to export…

Thanks for the honest feedback. The Excel message is just a test I did yesterday (taking the opportunity that the post was on the front page). I don't plan to tackle Excel head on at all. Just very specific excel use cases by financial institutions who totally abuse it.

Re: Ask HN: I built it, nobody came, now what?

#243

Post by d--b »

aabbcc1241 wrote: post #228

You do that to address your own itchy, so you don't have the pressure to promote it. Just use it as usual, if others find it useful, the popularity will grow organically. If it stays low profile, like many npm packages. Let it be. Unless you need to relay on others using it to pay your bill.

True, I need to keep that in mind. Thanks

Re: Ask HN: I built it, nobody came, now what?

#244

Post by AznHisoka »

_y5hn wrote: post #144
d--b wrote: post #34

Earlier quoted context omitted.

Yep! Glad to hear that you've experienced that problem. It usually ends up in hacky spreadsheets... Give it a shot if you face the same problem in the future :-)

If any hacky spreadsheet can do it, wouldn't developing with JS be harder and slower? Spreadsheets date from 70's and found their optimal form in 80's, so are much alike today. What does this tool provide beyond ie. OpenOffice? Do you use the tool for your own need, eat your own dog food?

Yep, I agree. I don't even think spreadsheets are that hacky. I prefer them to using JS, or any sort of coding. The visualizations Google Sheets provide are even more cleaner/visually appealing than the ones I see in most web apps too.

Re: Ask HN: I built it, nobody came, now what?

#245

Post by ReflectedImage »

samjanis wrote: post #128
ReflectedImage wrote: post #56

Yeah, I got to front page of hacker news but I'm only getting 80 unique users per day on my website. Through there are now also 200 unique bots per day. (Where do all these internet bots come from?)

Yeah, I got a load of those bots too. Not sure if it's the same as yours but from what I can see in the system log, most appear to look for vulnerabilies by trying to access places like '/admin' or '/passwd'. Thankfully a RewriteRule in .htaccess catches most of these although I can't help think if there's anything else I missed, security related. :\

If I had a thinkphp install it would of been hacked 10 times a day.

Re: Ask HN: I built it, nobody came, now what?

#246

Post by cixinlui »

d--b wrote: post #197
cixinlui wrote: post #192

Earlier quoted context omitted.

I don't think it is correct that "they don't have the choice". Reliance on EUDAs in Excel is a well telegraphed problem for literally decades. There have been lots of attempts to solve it, whether by apps which seek to replace Excel, addins or things which work with Excel, programming interfaces, changes made by Microsoft to Excel itself, alternatives such as Jupyter. You needed to know about these and understand to…

The main motivation is because "I was there". I've worked 13 years building EUDAs in Excel, with all sorts of addins and what not. I've been very good at it. Being the only developer on successful trading desks for years. Creating tools used by everyone at one of the best hedge fund on the planet. Literally everyone wants out of Excel but can't. Users are too addicted to the speed of iterations. Hiring UI guys to mak…

Do you think that the alternatives failed because they weren't 'spot on'? Perhaps they failed because it's a much harder problem than they recognized when they started out.

Re: Ask HN: I built it, nobody came, now what?

#247

Post by he11ow »

d--b wrote: post #167
danielmarkbruce wrote: post #145

Earlier quoted context omitted.

Can you be more specific on what you mean by "audited" ? What does "too many formulae" mean? Like what is a specific example? What does "not webable" mean?

audited as in "version controlled". In Excel, if you insert a column somewhere, all the cells move, and all the formulae change, so it's really hard to spot the differences between two versions of the same sheet. In excel, a lot of operations are "atomic", they work on numbers. If you get data in a table, and then want to operate on that data, you have to copy over the formula for each row in the table. Now, what hap…

In all of this 200+ comments thread, this is the first time you've clearly articulated examples of the pain to solve. Here's what to notice about these three comments:

In all three, you still want to ideally be using Excel. In a sense, these are three feature requests for Excel...and each of them is valuable:

-Getting a tool that instantly turns Excel into something made available online, that can be easily manipulated - that's great. (Not Google Sheets. Not Tableau. Not CSV. Excel) - A Diff tool for excel spreadsheets, or a version control tool for them - also amazing.

Your tool is none of those things. It demands so much upfront, and supposedly targets those who have neither the time nor the desire to spend that time fiddling with what is ultimately pitched as a sort-of-Excel-but-not-that-solves-maybe-one-problem-but-presents-five-new-ones.

Beware of the sunk cost fallacy. You've put all this effort into this and now it becomes how to patch it up into something someone will buy.

Re: Ask HN: I built it, nobody came, now what?

#248

Post by encoderer »

TenJack wrote: post #76
encoderer wrote: post #58

If your goal is to build a free tool that will be used and beloved by millions, you probably need to continue building on a deeply technical foundation and blog about it the whole way. It's ancient history now but a master class of this is when Damien Katz built and live-blogged Couchbase, like 2004. If your goal is to build an indie software business with hundreds or thousands of customers where you're making like 1…

What advice would you give if going for 10k-100k a month?

Some people here describe his landing page as “niche” but it’s not a niche at all. Rather, it’s a persona: ceo or cto with a team.

You develop personas after you know who your customers really are, and it’s probably not ceos.

Instead, he should find and focus on a true niche. Maybe people are commonly analyzing shopify store data using a tool like this. Find something like that, and just nail that use case. That’s a niche. Expand from there.

Re: Ask HN: I built it, nobody came, now what?

#250

Post by _y5hn »

AznHisoka wrote: post #244
_y5hn wrote: post #144

Earlier quoted context omitted.

If any hacky spreadsheet can do it, wouldn't developing with JS be harder and slower? Spreadsheets date from 70's and found their optimal form in 80's, so are much alike today. What does this tool provide beyond ie. OpenOffice? Do you use the tool for your own need, eat your own dog food?

Yep, I agree. I don't even think spreadsheets are that hacky. I prefer them to using JS, or any sort of coding. The visualizations Google Sheets provide are even more cleaner/visually appealing than the ones I see in most web apps too.

Yeah, but if you add complexity, they won't scale. So it's features like that that can make a new product stand out.
Post reply on HN