Mostly holes - part 4 : let's get this party started

After so much empty words, it was time to get in contact with reality again.
So, I've

• looked at the development of browser plugins, 
• decided for me, that the Chrome browser is a worthy first goal
• read some of the source, the smore team has used for their Clippy tool
• and with some sweat, scissors and much duct tape, I've got something to start with...

The project (if it should sometime be worth to be called that way) is hosted at Google Code and for curious minds, a first binary can be installed to torture your browser ...

• Just go the Options (Einstellungen)
• Go to Extensions (Erweiterungen)
• move the .crx file into the window and answer with "yes" when he asks to install it.
• ... the more comfortable way with the webstore costs some bureaucracy and 5 Dollars - so for now you have to do it manually ...

Mostly stone: Warum etwas Granit einem Garten ganz gut tut

Wir haben eine Doppelhaus-Hälfte und im Vorfeld gemerkt, daß diese vom Grundriss her fast immer gleich aufgebaut sind.
Auch beim Garten versuchen die meisten ähnlich zu optimieren:

Wenn der Garten eher übersichtlich als groß  ist, dann tut man sich erst einmal schwer, ihn mit Treppen, Kübeln und Abtrennungen voll zu stellen.

Glücklicherweise hat sich eine befreundete Landschaftsarchitektin unserem Streifen Lehm angenommen und etwas gezaubert,  das uns super gefällt.
So gut, daß wir auch anderen die Angst vor hohen Zäunen, Tonnen von Stein und schwindenden Rasen-Flächen nehmen wollen.

Uns ging es darum,

• den Blick von der Straße ins Haus etwas einzuschränken
• die Terrasse nutzen zu können, ohne den Nachbarn zu stören
• eine Abkürzung von der Straße ins Haus zu haben. 
• Ein bisschen Kräuter für die Küche anzupflanzen.
• Rosen für die Frau! Zumal sie in der Richtung bisher immer ein wenig zu kurz kam.

Im Detail

•  Die Hainbuchen-Hecke schirmt momentan noch nicht wirklich von der Strasse ab - aber im Frühling/Sommer sollte das anders werden - wenn sie erst einmal ordentlich blüht.
• Sie ist im Vergleich zu Tuja-Hecken einfacher zu schneiden, verliert jedoch für ein paar Monate im Winter ihr Laub.
• Die Hochbeete schirmen einen recht gut gegen die Straße ab 
• ... außerdem sind dort die Kräuter Schnecken-sicher 

•  Der Sandkasten war eine der wenigen Sachen, die wir zum Teil selbst gemacht haben - dann sieht man erst einmal, wie aufwändig es ist, ein paar Steine einzubetonieren...
• Auf der Rabatte (linke Seite) wird in nächster Zeit noch ein Sitzrost aus dem Terassen-Material geschraubt, sodass man sich auf den Rand hocken kann.

•  Der Platz für Tisch und Stühle wurde recht großzügig ausgelegt, sodass man auch mal ein Fest darauf feiern kann.
• Der Terassen-Belag ist Thermo-Esche, da sie eine ordentliche Alternative zu Tropenhölzern darstellt und recht langlebig ist.
• Die hohen Granit-Stehlen sind nicht nur wartungsfreier Sichtschutz sondern halten auch die Geräusche recht gut ab

• Trittsteine und Terrasse sorgen auch bei Schnee und Regen dafür, daß man zur Terrassen-Tür kommt.
• Die 2 kleinen Solar-Leuchten geben in den ersten paar Abendstunden noch ein bissel Licht, sodass man auch nach einem abendlichen Einkauf weiß, wo man hintritt
• Der Rosenbogen wird noch mit einer Kletterrose und einer Klematis überwuchert sodass es nicht gar so steinig aussieht wie bisher...

Wie schon gesagt - uns gefällt es wirklich gut und wir können die Vorzüge von etwas mehr Struktur nur jedem ans Herz legen...

Der Plan zum Nachbasteln

netterweise hat sich Christine Reess bereit erklärt, ihre Pläne auch zu veröffentlichen.

Und wenn's nicht ganz zur eigenen Situation passt - sie ist immer kreativ und ihre Dienste (als Landschaftsarchitektin) käuflich ;-)

Mostly holes - part 3 : To find good Baby-SITAs

In the last part, an enormous amount of work was presented
... and some little benefit, you might get back...

It is highly improbable to get a bigger number of people to
work on something that abstract and sacrifice a big portion of
their spare time for something that might never work out.

... sounds depressing? Don't worry ... we'll get them ;-)

Cheap tricks that might work once more...

In the dark times before I learnt about the joy of LaTex (not what you think - follow the link), I was using a student version of Microsoft Office 2000 along with its Office Assistant.

It wasn't, that I needed much help of it and seldom asked anything but I found the animations tremendously funny - just look yourself - I would recomend "Print" and "Empty Garbage".

SITA's chance might be to be around just like Link, the cat, was on my desktop.

Today that would mean a good written browser plugin - but in the end it should work more or less like back then...

Example of an interaction

• When the browser starts up, SITA might give you a short update what has happened on the net since the last time, you had a look
• It might fetch some RSS feeds, compare it with other users if they found it interesting etc.
• ... it might even do horrible stuff like looking if somebody has nudged you on facebook and stuff like that - whatever it takes to attract some babySITAs
• The user is mildly interested and starts his day with the news - he scrolls down and starts read an article
• SITA might look over his shoulder, see, that the focus is on an article about a big train accident and would stroll off into the net to see if some other user has commented on this or found additional infos
• It might "come back" and tell the user about its findings or - when he feels brave enough and hasn't bothered its owner for a while it might start to ask what he looks at and what the vocabulary means, it doesn't know yet...

It takes all kinds

• Some of you might already think "Oh, that sounds a bit too much - I just want to surf the net - don't you dare asking me things every 2 minutes"
• It would be a total disaster when the users would feel that way.
• Therefore SITA would have to be patient, uncomplainant and if it wants to start a dialog, it should do so in an unintrusive way like "Meeeow" - see the cat above with "GetAttention"
• The user on the other hand should have the opportunity to encourage SITA to do more or make him stay quiet for a while (and more often in the long run).
• It's a bit like the old Petz games or Black and White in that way, where you get other results relatively to the actions you treat the thing with.

Teaching an old Cat/Dog/Paperclip new tricks

• As long as the user doesn't have some real advantages, he won't install such a "code monster"
• Therefore we need THE KILLER FEATURE
• For the audience we target prominently (geeks) this might be something like "Sita - fetch me some comics"
• And Sita opens up all the comics, the user normally watches, in seperate browser taps - but not the ones, he has already read today.
• In order to do such a specific task with these constraints someone has to be quite smart about it
• At the start, this hasn't got to be SITA but a user, that has trained/programmed SITA to do that exact trick...
• When more and more of these tricks are implemented, the users will try to extract the things, that SITA has to keep an eye on.
  Once the tricks have a trivial part, that changes, and a common part with some more difficult logic a new trick can be implemented a bit faster and when a clear pattern emerges at some point probably by SITA itself.

Mostly holes - part 2: Educating SITA

It takes a whole village to bring up children.
It might take the whole internet to bring up an artificial intelligence.
=> therefore the name Synthetic Intelligence Taught by All.

The goal

The goal in the end should be to have a system, that can utilize a big knowledge base, give correct answers using it and as a glitter on the horizon be creative about it.

... and what you get when you start

Like a small child, the system might just start to repeat repeat the primitive sentences, you have just taught him when it thinks, it might be appropriate.
That is just about what the semantic web has accomplished for now. Links to word definitions in other statements - like in Wikipedia.

It would be nice, when some more knowing responses could be given.

Cheating

It might be OK - for a start - to pass questions, that are too complicated, to real persons working in the background, who answer them in the same manner as SITA would have done.
The system would watch closely and might answer the same or similar question the next time on its own...
And over time less and less human support would be needed... provided an enormous amount of wishful thinking.
Return of invest: You could start out with a (quite slow) system, that answers your questions - and it might be an overwhelming surprise when you get your first answer that no human has laid its hands on... and these cases would increase (hopefully) exponentially.

Showing SITA the world

A first approach might be a tagging of persons and things in pictures of news sites.
The tags would define the area inside the picture and sort the tag-description into SITA’s grammar along with a little sentence that describes the scene and hopefully contains all used tags.
Return of invest: You might get good tagged pictures since the described parts of the picture are pinned and the tags are set in context with synonyms and generalizations.

The Furby effect 

Next to a visual understanding of real life things comes an auditive interaction to make up something that might pass as human-like.
Return of invest: When a user trains the system, it has to his dialect and voice into account.
Over time it would get better at understanding and repeating these dialects so (after long years) a textual chat might be read to you in the voice of the other person, since their voice representation etc. are available in the cloud.