The ultimate tool: email

A few years ago many of us declared email bankruptcy, we all had in-boxes full of all types of crap and it was almost impossible to find the few useful messages in the sea of spam.

Progressively we moved our communication to other channels: blogs, twitter, facebook, various IM clients, other social networks, where we thought we could manage our communication better.

In the last few weeks I have missed a whole bunch of important messages which arrived on these channels: as a Skype chat (but I was away from my desk and didn’t notice it for a few days), as a Twitter direct message (I don’t check them very often) as a message on Facebook (I check those even less).

Meanwhile, my mailbox has started working again. With several layer of spam protection (before reaching me any email message has to go trough Spam Assassin, Greylist, Gmail filters Mail.app spam filters) most of the stuff I get in my mailbox is actually interesting, directed to me, and something I can manage pretty well with the tools I have (MacBook Pro, MacBook Air, more Macs at home, iPhone, etc).

Curiously this morning Loïc is declaring Twitter bankrupcy.

PS: of course, as soon as I posted this, gmail died.

Faces

I’ve been playing a little with iPhoto 09 Faces feature. It took almost 3 hours for my home dual processor G5 to go trough 15000 pictures and find all faces, but then it only took it a few minutes to learn how I look like and find me in several hundred pictures.

Me, me, me, me, me....

I don’t know how algorithms used in iPhoto compare to the ones used by security systems to track people in surveillance feeds. If they are somehow similar, we are kinda safe: while iPhoto did find a lot of matches, it’s probably less than 10% of the actual number of pictures with my face in them. It will probably improve in time by leveraging a larger number of models, but it still require quite a few human work.

What the software is very good at doing is recognizing that some groups of pixels are actually human faces. I lile the way iPhoto comes up with facewalls of faces, all cropped at the same size.

Hard not to wonder when Google will start running this type of code on their picture search site. With just a little “mechanical turk” contribution the precision will be stunning, and we will have to kiss another piece of our privacy goodbye.

PS: David Orban notes that google does have some basic face tools.

Sub-zero routing

To host our servers (and our clients’ sites) we have always tried to get as much quality as possible, choosing the best facilities in Europe and the US.

But when it comes to our office connectivity, some time ago we decided to go for quantity over quality: at more or less the same price we could have had a super-safe 2mbps DSL connection, backed up on ISDN lines and with “serious” Cisco router, or a very fast 10mbps ADSL+ line, with a cheap leased router (the type that feels cheap and empty when you hold it, probably only because Cisco puts some useless iron as ballast in their boxes). We went for the fast line and we have been very happy with it for the last six months.

Until last week, when the little cheap router decided to forget its configuration and die. The ISP first blamed the DSL line, then (after waiting for 24 hours the visit of a Telecom Italia guy who said that the line was fine) offered to send us another router… in a few days.

Luckily we had a spare cheap router in the office which we plugged in to keep us afloat, we kept having problems, the router would restart every few minutes, making our life miserable.

So we borrowed another router from a local shop, a cool Linksys box, and I wasted 3 hours trying to set it up with the help of our ISP. No way. It would simply not work with our configuration (we have a bunch of IP addresses to host some dev servers).

Somebody suggested that the cheap self-restarting router could have had a heating issue, so last night I found some longer Ethernet cables and moved the router outside the server room window.

sub-zero

When this morning on the radio I heard that the temperature in Gorizia had fallen to -3C°, I thought that the little router should have been perfectly happy and pretty cool. But as soon as we started making IP traffic this morning, the line started dropping again.

Finally we borrowed yet another router, this time a Netgear, and in only 10 minutes we were up and running again.

No traces of the new router from the ISP yet.

Anathem

I’m reading Neal Stephenson’s latest book, Anathem. I like Stephenson: Snowcrash and Diamon Age are classics and Cryptonomicon is one of my favourite novels ever. Things started to get a bit complicated with the Baroque Cicle, and Anathem is even more… uh…  intense.

Stephenson’s novels have a strange effect on the way I think. It’s not strictly related to the stories (actually in the first 140 pages of Anathem are more about setting the stage then telling a story), but they way arguments are treated is triggering all sorts of lateral thinking, in particular related to software architecture design. Don’t ask me why… it just happens. If you are developing software, you should give it a try.

Also the name of our company is related to a Stephenson’s book: I read a few times the term “vectors” (I think it was in Cryptonomicon) and back then every company had a “e-” in their name, hence: evectors.

Anathem is set in a different world, with a different history, different traditions and its own dictionary. Everything is different, yet everything sounds very familiar.

One of my favourite terms from “the dictionary, 4th edition, A.R. 3000“, is:

Bulshytt: (1) In Fluccish of the late Praxic Age and early
Reconstitution, a derogatory term for false speech in general, esp.
knowing and deliberate falsehood or obfuscation. (2) In Orth, a more
technical and clinical term denoting speech (typically but not
necessarily commercial or political) that employs euphemism, convenient
vagueness, numbing repetition, and other such rhetorical subterfuges
to create the impression that something has been said. (3) According to
the Knights of Saunt Halikaarn, a radical order of the 2nd Millennium
A.R., all speech and writings of the ancient Sphenics; the Mystagogues
of the Old Mathic Age; Praxic Age commercial and political
institutions; and, since the Reconstitution, anyone they deemed to have
been infected by Procian thinking. Their frequent and loud use of this
word to interrupt lectures, dialogs, private conversations, etc.,
exacerbated the divide between Procian and Halikaarnian orders that
characterized the mathic world in the years leading up to the Third
Sack. Shortly before the Third Sack, all of the Knights of Saunt
Halikaarn were Thrown Back, so little more is known about them (their
frequent appearance in Sæcular entertainments results from confusion
between them and the Incanters).

Back to blogging

Scoble writes:

When I go back and look at my blog back in 2004, for instance, it looks a whole lot like Twitter. Short item with a link.

It’s true for me as well, looking at my first posts, most of them were one sentence, pretty much what we do today with twitter. Back then posts had no titles and were organized by days.

This was what Radio UserLand was offering, this was the way Dave was blogging, and we were all following. Now, almost 7 years later, I spent the last 2 weeks in the Bay Area meeting old friends, some of the smartest people I know, and maybe not surprisingly quite a few of them were Radio users in those early days.

While it is true that microblogging has found other tools and spaces, I am realizing that I’m missing a space for longer posts, that take a little more thinking (but that require more time to write). Even if I am not been writing on this blog for a while, I still compose posts all the time in my mind, so I guess the time is right to try to get back to blogging.

Thanks to one of the 25 nicest people of the web ;->

I’m back

We are just back from a 2 weeks trip to San Francisco. The reason for this trip was exploring the possibilities of developing a business in the US around our aggregation/publishing product, PagesPlus.

We met old and new friends, agencies, potential clients and investors, and the feedback was very positive, even better than I was expecting.

It sounds like the mix of tag-based RSS aggregating and WYSIWYG page editing is very interesting for publishers and media companies (these are our core business in Italy) but also for PR firms, web agencies, social media projects and as a platform for behind the firewall applications.

I’m very happy to start this new adventure in the software business with great partners and friends, I’m already planning my next trip to California and we owe tons of demos to a lot of people we met. Stay tuned.

Evectors PagesPlus

For the last 10 months at Evectors we have been working on a new application which we are calling PagesPlus.

It’s not strictly a content management system: most of the content is not authored within our application, it’s more a content presentation system, or if you like a tool to manage content flows in an editorial environment.

The core of PagesPlus is an aggregator which can digest any form of RSS/Atom and uses tags to organize everything it aggregates (it’s loosely based on the concept that, together with Matt Mower, we developed for our K-Collector application back in 2002). The aggregator supports tag schemas, meaning not only that, for example, it can tell the difference between a topic-tag and a category-tag, but also that you can create your own schemas to address special needs.

If the aggregator manages the back-end, a WYSIWYG application allows you to manage the front end. We are using again the metaphor of “plus” buttons: you click on a plus button to add content in a specific area of the page, you choose what type of element you want to add, and seconds later the page is updated.

The third part of the application is a “widgeting” platform. With this latest addition you can detach any module from a PagesPlus page and insert it in any other page on the web. The system does all the necessary CSS/HTML/JavaScript magic that will guarantee your widget to look the same everywhere you paste it.

Just to give an idea of what it can do, using PagesPlus in the last months we have have developed these sites:

Using PagesPlus to create site with very strict specifications has been a very useful exercise, forcing us to eat our own dog food and putting us in the shoes of designers, editors, writers, publishers.

I must say I’m very happy with the results. Not only we have been able to put together a killer team, but we also ended up having a very good product (whith an even more impressive list of features on our roadmap).

The plan is to release the application with some open source license (still trying to figure out which one to pick) and let a thousand flowers bloom.

Hoping to have a little more time available in the next few weeks, I’m planning to post here the description of how the system works, maybe a few screencasts to show how the real thing works. Meanwhile, if you have any questions, feel free to comment below or to ping me.

On my way back

I’m at the airport, too early as usual, on my way back from Reboot10.

It’s almost trite, but it’s worth repeating: Reboot is fantastic experience. It’s not only the great talks (unfortunately some work related issues forced me to sit in a corner of the public spaces if not in my hotel room for quite some time and I have lost some very interesting ones), most of all it’s the people.

This conference attracts a crowd with an incredible percentage of very, very smart people. It’s exciting and humbling, a great experience. It’s definitely the best event of the year.

The list of interesting people I met is just too long to report (I would forget too many names), and the best part is that it’s not only the usual suspects, I have met for the first time a whole bunch of persons that I’m really looking forward to meet again.

Thanks everybody for these extraordinary days.

Reeplay.it

Reeplay.it is a company I have been working with for a while, we presented the concept last year at Le Web 3.0, and finally last week we opened to the general public: now everybody can subscribe and use it.

Here’s a short introductory video:

The basic idea is simple: bookmark your favorite videos and play them back on any device. We’ll take care of storing, transcoding, distributing, sharing, etc.
 
There’s a ton of additional features in the pipeline, but feature-wise I think we have reached critical mass: it’s useful enough to get the first users on board and start collecting feedback.

Hello World

Welcome to this new blog.

All my old stuff has been saved and it continues to be available at the old urls (well… at least this is what I’m trying to do).

But from this post on I’m continuing with a new tool, a new template and hopefully some new ideas.

I wonder if there’s still anybody out there.

:-)