5a build a better google alerts

0 views
Skip to first unread message

JakeCarpenter

unread,
Dec 28, 2011, 2:17:07 AM12/28/11
to Startup or Shut Up
5a Let me throw one out: I think someone could build a better google
alerts, and charge for it. Everyone I know in tech uses it, and I
don't feel like google is investing in it, certainly not for very
specific use cases.

^Yeah, it works, but there is SO much room for improvement.

^ This is an interesting idea. So are you thinking maybe trying to
build it ontop of google and then putting additional filters in?-I
don't know about that, but i like the direction we're going.

I like this. If we are going for specific use cases, would we be
working directly with the people who want this. Eg. some finance
company wants custom alerts on aluminum trends or whatnot, and we
charge it as a service. I have no idea about the business end, just
throwing this out there

^ I'm thinking we may want to start off "small" building on top of
google's current infrastructure. Only because unless we can find
services that will alert us when say the price of aluminum hits some
amount we have to sit their constantly polling some service.

^not to play buzzword bingo, but the integration of social/local
could help it a lot. When the price of aluminum hits X, someone is
bound to talk about it. A more appropriate case would be the restarant
owner with an alert for his biz name, but probably also cares about a
competting restarant in his area getting press or social accolades
locally. or a VC (they're notorious for this)

^ We could apply filters so they could specify to only get a
notification if a critical comment was made

^im not sure sentiment analysis can be trusted at all< Are there
alternatives that acheive a similar result? We just need to be better
than google alerts, not psychic.+1

Ok so what don't we like about google alerts?

Lots of false positives+1

No way to specify exactly what sources?

No way to limit geographically? +1 I started to make a newsfeed
for sales territories

Limited to a specific term, little more than search and notify.
(lacks intelligence?)

This isn't the answer but: lots of people have google alerts set
for their own name, or company, so they're aware of everything said
about them - what can we do with that? There are companies that do
"media monitoring" I.e. who picked up your press release and how many
people saw it, but I don't know if that is done for the long tail,
except obviously for social media mentions.

^how does a project like this dovetail with our strengths?

A business owner could get alerts for 'keyword' news items within
their zip code or other sales area. chrise

JakeCarpenter

unread,
Dec 28, 2011, 2:38:23 AM12/28/11
to Startup or Shut Up
This project is one that I would like to work on.

I can code PHP/JS/HTML/CSS/MVC, and do some biz stuff. Would like to
learn marketing. Design is not a strength I possess.

Jasdeep Hundal

unread,
Dec 28, 2011, 4:38:13 AM12/28/11
to Startup or Shut Up
I like this one as well. As for my skills, I'm mostly a coder. Pretty
agnostic regarding specific technologies, I can pick up things more
than fast enough. That being said, Python is what I prefer (also have
some experience with Django), but I've dabbled in lots of different
things.

On Dec 28, 2:17 am, JakeCarpenter <jakearthurcarpen...@gmail.com>
wrote:

Chris

unread,
Dec 28, 2011, 10:55:48 AM12/28/11
to Startup or Shut Up
I am interested in this as well. I can provide data mining, api
support, front end, and geographic data integration/analysis.

I think a google alert tailored towards business intelligence - for
small business and sales people, to alert them of changes or
opportunities in their business territory, would be a valuable feature
we could get subscribers for.

Most businesses and sales people have a geographic 'territory' (this
area can consist of counties, zip codes, census tracts, or a
combination of these). If the user can define their territory and use
this as a filter combined with a list of topics (business,
environment, regulation related stories) and keywords (their company,
competitors, grants, etc.) this would provide very useful reading.

What do you think? (I included a smaller version of this idea in the
etherpad)

Jesse Pollak

unread,
Dec 28, 2011, 11:05:54 AM12/28/11
to startup-o...@googlegroups.com
Interested in this project as well, although I have no experience with big data processing. Seems like a great idea though.

I place myself in the coder category and have the most experience with Ruby/Rails. I'm open to picking up any language or framework though and consider myself a fast learner.

Excited about this,

Jesse

Chris

unread,
Dec 28, 2011, 11:48:26 AM12/28/11
to Startup or Shut Up
Worked on a newsfeed map viewer last year:
http://newsfalcon1.appspot.com/maps/view

Jake Carpenter

unread,
Dec 28, 2011, 12:42:14 PM12/28/11
to Startup or Shut Up
Great, this project looks like where I'll hang my hat. I'm fine with
using Python, but it may take me a while to get up to speed. Should we
use AppSpot? Heroku/Django? Plain Ole Django? We should see if we can
recruit a designer from when of the less voted for projects. I'm going
skiing, but I'll be back this evening.

-Jake

Jake Carpenter

unread,
Dec 28, 2011, 7:43:59 PM12/28/11
to Startup or Shut Up
Skiing was great, but now its time to get to work. I think we've got
enough buy-in to start as a group, so I've started our own etherpad:
(http://notes.avaaz.org/KRa20nRIhg) for group collab for anyone. I'm a
little concerned that we don't have a desginer, but maybe we can
squeak by with a themeforest site template until we recruit one.

-Jake

On Dec 28, 8:42 am, Jake Carpenter <jakearthurcarpen...@gmail.com>
wrote:

Jasdeep Hundal

unread,
Dec 29, 2011, 3:46:31 AM12/29/11
to Startup or Shut Up
Should we try to set some sort of meeting time (say once a week, maybe
a couple in the next two days though)

On Dec 28, 2:17 am, JakeCarpenter <jakearthurcarpen...@gmail.com>
wrote:

Jake Carpenter

unread,
Dec 29, 2011, 12:56:08 PM12/29/11
to Startup or Shut Up
I think that we should strive for 2 times a week, at least. I can be
available almost anytime.

-Jake

Chris

unread,
Dec 29, 2011, 3:38:24 PM12/29/11
to Startup or Shut Up
Twice a week is great. Even google chat or thru etherpad will be
needed for a 2 week project imo.
Let's chat soon so we can talk live about the project, tasks
breakdown, logistics, etc.
Would Skype be cool?

-Chris

On Dec 29, 9:56 am, Jake Carpenter <jakearthurcarpen...@gmail.com>
wrote:

Jake Carpenter

unread,
Dec 29, 2011, 3:44:56 PM12/29/11
to startup-o...@googlegroups.com
I'm fine with skype, but I have pretty slow net connection (verizon 3g FT...W?), so video is likely out.

Jake Carpenter
(907)321-8693
ja...@jakecarpenter.com
P.O. Box 34783, Juneau AK 99803

Rory W.

unread,
Dec 29, 2011, 10:27:01 PM12/29/11
to Startup or Shut Up
Interested in this project,

Coder (ruby/rails) looking to improve and hone my current rails
knowledge.

Design isn't my forte.

Have a language/framework been finalized?

On Dec 29, 3:44 pm, Jake Carpenter <j...@jakecarpenter.com> wrote:
> I'm fine with skype, but I have pretty slow net connection (verizon 3g
> FT...W?), so video is likely out.
>
> Jake Carpenter
> (907)321-8693
> j...@jakecarpenter.com

Jake Carpenter

unread,
Dec 29, 2011, 10:37:10 PM12/29/11
to Startup or Shut Up
Hi Rory,
You're welcome to join our project, the more the merrier. At this
point, we haven't finalized a language. If you add your name to the
list with your email/skype name, we'll include you on further group
correspondence. I'd encourage you to reed through the message in the
Startup Or ShutUp [SuSu] group if you haven't already. I think we'll
all be in and out of the etherpad over the next 18 hours or so getting
ready for our first skype meeting tomorrow afternoon.

-Jake
Reply all
Reply to author
Forward
0 new messages