Mark is starting to write a bit more about BrainPark. Finally I’m a freedom fighter!

Oh, by the way, apparently I work for one of one of Canada’s hottest innovative startups. Is this considered bragging?

Neil posted today about restaurants and the challenge they have getting real feedback, assuming they’re one of the rare few who actually want feedback. The way we’ve been tackling feedback in BrainPark is to try and hide it from our users. Hide sounds a tad malicious but all that I mean is make feedback part of the application instead of soliciting feedback from users with forms and other garbage.

A somewhat trivial example, let’s say you’re building a code search tool for your intranet. You want to know when you’re search algorithm sucks and when it works so that you can improve it based on hard facts. What do you do? Add a “Did these search results help you?” form right? Ewww!

Build the feedback into your application. In this case, add an X beside each search result that allows the user to make that result go away. For the user, they can refine and shape the results to give them more value and reduce the noise. For you, you can now create reports that show your team precise examples, with real data, of where your search is failing.

If you could somehow add a similar feature that allowed the user to say ‘I should have saw this in my search results’ then you’re in even better shape. Now all your design/dev team has to continually reduce the amount your users have to ‘fix’ their search results.

Back in my personus days before the dotcom blowup we built several online canadian pharmacies. I can’t recall all the names but I’m pretty sure they’re all dead now. Ali and his team at well have been working hard to fill that void and have pulled it off and then some.

Today Jonas and Ali announced that well.ca just landed the “largest investment the Maple Leaf Angels have made to date”. A big congratulations to Ali and his whole team. Our BrainPark crew has been lucky enough to share well.ca’s office space for the past few months and it’s been great getting to know them all. I’m just surprised to see they’re actually getting work done with all the damn fun they have there.

We’re after developers at BrainPark, description and more details here. We’re a startup operating in Guelph building primarily on a python based technology stack which I think offers some uniqueness. Experience with python is nice but not required. It’s all about you, skills can be learned. Feel free to contact me directly if you’re interested.

The MultipleIndexes wiki page may, or may not, contribute to the previous post…

“There are various strategies to take when you want to manage multiple “indexes” in a Single Servlet Container”

After previous experiences with lucene.net, which weren’t all bad, I’m enjoying experimenting with Solr.

“Solr is an open source enterprise search server based on the Lucene Java search library, with XML/HTTP and JSON APIs, hit highlighting, faceted search, caching, replication, and a web administration interface.”

It was very simple to get an instance running locally with jetty and I had a test python client connected and searching using simplejson within minutes.

Here’s my question. I’m curious how applications are using MultiCore on Solr and what limits they’ve been able to take it to.

Let’s say you’re building an application like SalesForce where you have a lot of clients who technically have no need to search across each other. As well, each client may have multiple indices, let’s say 3 for the sake of having a number. Do you?

1. Use MultiCore to dynamically generate unique indices for each new client resulting in creating and managing 3*(number of clients) indices.
2. Create 3 potentially massive indices which will clearly impact search performance.

Assuming Solr and MultiCore can handle #1 then it seems the clear winner to me. Anyone??

I have a new favourite, and one of my most used, applications. First, a quick history. I don’t like using a mouse and I don’t like menus, especially start menus. To avoid start menus I used to start most apps from a shell by creating a multitude of alias that I’d have to remember.

After that I moved onto using katapult which was better but flaky, likely because it was built for kde not gnome.

After that I moved onto deskbar-applet which I still use a bit today. I believe it comes on the ubuntu base install so you usually have it from the start.

I’m now using gnome-do for all this and it’s the slickest one yet…

“GNOME Do allows you to quickly search for many items present in your GNOME desktop environment (applications, Evolution contacts, Firefox bookmarks, files, artists and albums in Rhythmbox, Pidgin buddies) and perform commonly used actions on those objects (Run, Open, Email, Chat, Play, etc.).”

It’s inspired by, which all the above are, Quicksilver. Gnome-do has done the best of them all. If you’re on a linux OS and haven’t yet, give it a try.

Just a quick thanks for everyone who came out last Wednesday, it was another great event. My quick poll says we had around 50 in attendance which is just about right. Keep an eye here or on our google group for DemoCampGuelph6 which will likely be sometime in July.

Your very very last reminder about tonight’s DemoCampGuelph5. It’ll be a fun night as always, say hi if you make it out.