Comment filtrer les informations confidentielles sans dupliquer les sources ni la documentation ?
Lors de l’ajout d’information : répétition de travail de saisie : perte de temps et source d’ erreur.
Lors de la consultation d’information : risque de ne pas avoir la version à jour.
Exemple : Une liste de produits gérée par le département engineering de la société.
Solution rationalK :
Une nouvelle fonction a été développée dans rationalK pour faciliter le partage d’information d’un département vers les autres départements de la société ou vers l’extérieur de la société (clients et fournisseurs). L’objectif est de ne maintenir qu’une unique source de donnée pour éviter les erreurs de saisie, les erreurs de version et surtout accélérer le travail de saisie souvent rébarbatif. Le département propriétaire des données saisie les informations dans son fichier Excel déjà existant. Il n’y a aucune habitude à changer pour les ingénieurs. La deuxième étape consiste à écrire un petit fichier de réglage dans lequel sont définit : les colonnes et les lignes à afficher.
Et voilà ! RationalK crée automatiquement un url pour accéder directement à cette base de donnée restreinte et sur-mesure. Le contenu de cet page url peut-être protégée par un mot de passe ou laissée public ce web en fonction des besoins. Il est possible de créer plusieurs url en fonction des destinataires.
A chaque modification, le contenu des pages externes sont mis à jours automatiquement et de manière transparente.
Certains détails techniques sont masqués. Le choix des colonnes à afficher se règle dans rationalK.
3) Url : http://masociete.com/external
Seuls certains produits choisis sont affichés et les informations confidentielles (colonnes) sont masquées. Les lignes à afficher sont filtrées par le mot clé Website dans la colonne filtre.
Pas de saisie en double
Aucune connaissance en base de donnée nécessaire (l’utilisateur doit juste savoir utiliser excel)
Mise en forme multiples et sur-mesure pour les utilisateur externes au département engineering
Mise a jour immédiate
Aucun changement d’habitude
RationalK est un logiciel de gestion de la connaissance permettant de capturer le savoir des employés de la société et de le mettre à disposition de toutes les équipes afin d’améliorer l’efficacité au travail et diminuer les frustrations.
I had a very interesting discussion with the Quality department this week.
I was presenting them rationalK, the knowledge management tool.
They did not accept the idea very well… They asked me : Why do you want another tool to manage the document, we already have the quality document management system ?
I think they misunderstand the difference between a knowledge management tool and their quality system to manage the company’s documentation.
Here was my answer :
The two tools are fully compatible and have totally different purposes.
Quality Document Management System
Ensure consistency in the company processes and the corresponding documentation
Associate a document to a process step
Guaranty an access to the most up-to-date documentation
Prevent sensible information from behind freely accessed within the departments
Most of the documentation inside the quality department are procedures or instructions
Make sure that documentation follows the company’s guidelines in terms of content and form
Help the worker to do their everyday work
Allow to freely work on official documentation without disturbing the quality processes. Documents are considered as draft in rationalK as long as they are not approved by the quality department
rationalK allows to browse within internet-downloaded documentation, within emails, notes, international norms, patents, project documentation, best-practises and even people’s competencies. This is definitely not possible (and wished) in a quality document management system
Document in rationalK are living document. They are constantly updated with new knowledge. Their accuracy depends only on the authors competencies
RationalK has its own access rights policy to easily spread knowledge within the company departments
In rationalK we have functionally that can link a document with the related official document from the quality department document management system
How they could work together
Let’s say a team of five engineers is using rationalK to manage their knowledge.
Every time that one the guy is doing something related to a project, he naturally stores the document in the project folder. Fine. If he thinks that this knowledge or solution could help the colleagues solving their own difficulties (what is often the case), he add an entry in rationalK. He can type in some tags, a brief description of the solution and of course the link to the file in the project documentation. At this stage we have exclusively used rationalK and no quality document management system ! A co-worker will never upload all the project document to the quality document management system. Thanks god !
Now if the problem-solving method appears to be more and more used within the team, the team manager may find useful to add it as a standard for solving a typical problem. Fine.
Now and only now, we can write a process documentation to be sent to the quality department. They review and approve it and then store it in the quality document management system. This may takes weeks ! In between the solution has improved. rationalK still have the most up-to-date best practises while the quality department system has a general document describing the methodology but the the details.
As a conclusion, I would say that rationalK is a tool to help people do their work according to the quality procedures stored in the quality management system. Quality documentation is the one that define processes on how to work. I really do not believe that any coworker could conduct a project by using exclusively the content of the quality document management system… But combining the two tools is the most powerful way to achieve efficiently the best results.
When I would like to test live my code : I push my dev files from my dev folder to beanstalkapp
beanstalkapp keep a version of all my work help my colleagues to review the code.
beanstalkapp is set up to automatically deploy my dev file to a ftp server.
I use ovh and a public dev folder http://thomasdt.com/myproject_dev
This is a php-sql project so :
I maintain a duplicate of the public database
If you are browsing the application through :
http://thomasdt.com/myproject, the php files are pointing to the live database
http://thomasdt.com/myproject_dev, the php files are pointing to the test database
When I would like to push the dev to the live site. I commit my files with the keyword: [deploy: production] and the files are also pushed to the production ftp server.
If I need to alter the sql database structure, I do it in a sql.php file rather than in phpmyadmin. This sql.php file is automatically triggered (like a web hook) by beanstalkapp just before deploying the file to the development and production server.
Things I could improve
On windows, I am pushing files to git with Git Bash and this is not optimal. I could find a better IDE with integrated git.
It helps the note taker to remind himself about important facts discussed in a meeting or during a presentation.
You can make them way more useful if you convert them into organized knowledge.
Under organized knowledge I mean a document recorded in the company database.
A good practise would be to convert every items of your freshly-taken notes into :
An assigned action with due date
An update of documentation like a specification or a handbook
An update in the project decision log
An update in the risk (or opportunity) register
A short communication to colleagues
A planning update
A change in the software
Whatever you want but something else than a hand written document than no one will use
Don’t wait too long to do this. Try do do it live or at least within one week period.
Doing it quickly after the meeting prevents :
People to wait too long for your feedbacks
Having to convert tons of notes at once.
This seems obvious but sometimes it is useful to recall obviousness.
As I go on with the development of the Rocket.pm and especially the Document Managing System (DMS) function, I came across the definition of a key role in knowledge management : expert knowledge manager.
Let’s assume that I have a document database in which documents cover different fields of expertise (like steel materials and ceramic material). I can define one (or more) colleague(s) as an expert in a given field; For example, Bruce is defined in the system as an expert in ceramic materials.
The document creator (let’s say a colleague from the ceramic material engineer department) has not been able to fill in every form field (or attributes) when submitting the document. I really don’t want to make all field mandatory because it may make the document creator reluctant to add its document to the database. That is why, the system let him submit its document with an incomplete form and as a consequence the entry in the database has some missing information.
Here comes the Expert Knowledge Manager in action.
The DMS will regularly ask the right Expert Knowledge Manager to fill in the blank fields. If a document cover the ceramic material subject then Bruce will be asked to improve the database entry. In other words, the expert knowledge manager make sure that all documents in his field of expertise are well documented and can be found in the database.
The philosophy behind is always the same : a DMS will work only if all documents are properly defined and all attributes are correct.