From ICISWiki

Jump to: navigation, search

Day 1: October 13, 2010

Venue: M.S. Swaminathan Building, International Rice Research Institute, Los Banos, Laguna.

Attendees:

Guoyou Ye (PBGB-CRIL)
William Eusebio (PBGB)
Grace Capilit (TTCGRC)
Victor Alcantara (PBGB-INGER)
Warren Constantino (PBGB-CRIL)
Dec Arreza (PBGB-CRIL)
Ching Habito (PBGB-CRIL)
Aien Capistrano (PBGB-CRIL)
Luisa Tabada (PBGB-CRIL)

INVENTORY MANAGEMENT SYSTEM

Guoyou: You should have an interface for modifying the ini-file. Breeders must not need to open the ini file to edit its contents
Grace: There should also be a filter for the viability
Warren: Filtering by viability is already done
Guoyou: Inventory should not be the starting point. It should be part of the process but not the starting point.
Grace: Should be based on the data not on the amount.

SEARCH GERMPLASM

William: How to search for seed stocks.
Warren: Use the Search Germplasm interface of ICIS Lite. It will show the equivalent interface for gms search including the seedstocks. Just tick the Show seed stocks available checkbox
William: Do not display the search results of germplasm that do not have a seedstock.
Guoyou: It's hard to locate those icons. Make it menus instead of icons.
Warren: We can customize the menu and icons.
Guoyou: This customization should not be done by users but by your group or data managers. Your default menu should be the menus that are common to most users not all the functionalities.
Warren: We are getting the workflows of different OUs so we can customize the menus for them.
Guoyou: This customization could be part of ICIS installation.
Grace: Menu Captions should be standardized.
William: We can do that
Grace: When they present to the a group of users, menus should have already customized.
Warren: We'll do that.
Grace: Is that searching also available for incoming seeds?
Warren: Yes.
Grace: Is it possible to search for different location per germplasm?
Warren: Yes
Guoyou: What is our objective for doing this? Why do we do this?
Warren: Because when users get incoming seeds we need to link them with the information attach to them. That is why we do the search.
Grace: Can you demonstrate how to do it for incoming seeds?
Warren: (launches Input Germplasm interface.)

LOOKUP ENTRY

Grace: Does it have to be displayed by ID?
Warren: You can choose display the name of the location too.
William: The user should not be able to manually change the sheet depending in their access.
Grace: If you put the curser in the id, it should popup the description for that id like name.

INPUT GERMPLASM

Grace: What is the female parent field for?
Warren: This is for the pedigree. If the system didn't parse the pedigree correctly, you can edit using the female male parent.
Grace: What if it is not a breeding line?
Warren: It should change. It is just a label.
Grace: Female should be the group
William: Is it working already? I want to change it as IR 64. Should make it work. Sometimes, the entry looks like a derivative but it's not so user should be able to make the change.

Coffee break

PARENT SELECTION

Guoyou: Should have scores. If there are too many studies, users should have a clue how to select a study. If not, it would take ages before a user can retrieve data
William: Probably, you can specify the type of study
Guoyou: For any study, should have validation to check for the essential requirements like location, year. The user must not be able to load the study if the information is not sufficient. Do we have to modify the schema to do it because

even if we set standard protocols before uploading a study, you cannot expect all the users to follow it

Warren: It can be done easily. We'll use the study table.

GENERATING CHARTS

Guoyou: Evaluate the tools being used. Results should be fast. Consider R. This function should also be part of the Parent Selection. Breeders should also be able to access basic functionalities like standard deviation etc.

GENERATE CROSSES

Guoyou: Can we manage the lists? the longer the list the more time it requires when retrieving. So when ICIS has been used for many years, lists are accumulated. Can we put a cleanup?
Grace: We should not delete the lists cause sometimes we need to trace back. Just put a list status like "inactive".
William: To save time, only load the lists of a folder when a certain folder is clicked.
Guoyou: How can I double check? The list should not auto-close after selecting the parents.
Grace: How do you cross the parents?
Guoyou: You should have more options in crossing. Should be able to handle duplicates.
Grace: Put a check if cross has already been made.
Ching: If you are planning to port everything from setgen in icis lite, you should start using the ini keys of setgen
William: What about reciprocal cross?

Lunch Break

Grace: What if I want to get the female parents from one list and the male parents from another. What if i have already selected a male and female parent but its not yet the final parent, i want to add another one.

Goyou: If checking for duplicates, should have the option, remove duplicates from the list.
William: We should allow using the same material but from different sources
William: For convenience sake, have the sources should also be made visible (the cross)
William: Need to hide the list as part of the source. select only the entrycode in the naming convention (that should be the default)
William: Have an option to save the preferences/default values you have entered in the naming convention windows so you don't have to repeat the values you entered.
william: can you adapt the setgen's form? it is quite convenient
Grace: an option to save your work if you are not yet finished generating crosses. save the last state.
Goyou: is it needed?
Grace: yes, sometimes you have to leave and do something else even if you are not yet finish generating crosses.
Grace: if you cannot display multiple list in the middle, can you just show the crosses being done.
: list for selecting male
: list for seleting female
: list of resulting crosses
Aien: can we remove the one entry in the list of female parents/male parents?
Warren: we can do that. it has been implemented already in other icis lite forms. we can implement it also in this
Warren: we can place a button if you want to see the resulting crosses
Grace: but that is how make the list. we see the sources and we see the list of crosses so we can easily delete if we make an error.
Warren: whenever you click the generate crosses it is already registered in the database
Guoyou: in my opinion, we should not have to many buttons here (lower part of generate crosses window with germplasm, names, attributes tab)
Guoyou: but that should be part of quality control, wrong data must not be entered in the database
Guoyou: so is this form good enough?
Guoyou: Dont write the Generated Crosses List right away to the database, Preview it first and ask the user later on whether to save it or not
William & Grace : Add the list to be opened directly either to the Female Parent Set or Male Parent Set
Guoyou: Decide which way to go, go with one list for female and male.
William: Search text box (Name & Entry Code) in the Generate Crosses form
William & Grace: In crossing, we select one female then cross it to 3 specific male germplasms then we use another female and then cross it to the rest of the set of Male germplasms
William: for the meantime, it should cater all how we actually do it
Guoyou: if we will always adapt it how you actually do it, it would be difficult for the icis team to make this software generic. we have to prioritize the flexibility.
Guoyou: single pair, NCII
Warren: we already have the NCII, it is the single pair that is not yet included
Grace: so how are you planning to implement the single pair
Guoyou: <setgen interface. pointed unused form (BTC) in setgen matrix crossing incoming rice seed>
Guoyou: most important thing is to allow user to check first the crosses before committing this in the database

BARCODING

Guoyou & Dec: save the barcode mapping so you don have to map it already all over again
Warren: it is being save already
Guoyou: place a default barcode template but user will have option to change the template
Guoyou: 99% of the time, user will use the default. user should only do the mapping if something has to be changed in the barcode label template=regular breeder should be able to generate barcode labels after a few clicks only, not so many clicks/steps
Grace & William: x button = interactive generating crosses
Guoyou: is sorting already part of that icis lite release?
Warren: not yet but it can be easily done

break

ADVANCING LINES

take note: - is different from B-

Guoyou: what taking its slow (pressing the advance line button to load the list explorer window)

: again, you need to have default settings for this form.
: if you dont want to have defaults, the first time the user does it, save his work as default
: why are there are lots of creation methods, are they all relevant?

William: just show the creation method for derivative and maintenance. do not show all the methods
Warren: how do we know that we are using SF?
William: if it is rice, then it is SF. order should be designation -> prefix -> suffix
Grace: should be loaded in the study
Ching: are you able to pick up data stored in a study when selecting for no. of plants selected, etc.
you mentioned that the same form is used for - and B-. you should be able to type in the no. of plants selected or pick the name of the study factor/variable from which to take the data.
Grace: delete in the list and the delete in germplasm should be different
Guoyou: should be able to go back if you want to edit your entries
Ching: i think there should be more options like range.....
William: are you mixing the - and b-?
Warren: im using the same form
William: this one has more flxibility cause its asking the number of plants to advance but it think you should limit it just to that. B- should be separated.
Ching: you can still use the same form if you want but hide other functionalities that is being used for b- when you are doing -

Day 2: October 14, 2010

Continuation of Advancing Lines/Generate Crosses Discussion

Grace: what about seed increase
Warren: seed increase is the same as advancing lines, right?
Grace: yes but we haven't seen what we need yet.
William: we had to set some default that we need to do in seed increase
Grace: is your target user from the genebank?
William: yes
Guoyou: you use single plant selection here. but you have to specify the number of seeds when collecting seeds (crossing).
have one column for the number of the seeds. from F1 to F2, you use single plant selection but its not typical for other generations
Grace: for genebank, we assign a different designation, we include the cropyear when it was planted.
Guoyou: then you have to set it in the naming convention windows
Grace: it can be easily done if you will include the mgid in the options to choose from. for the suffix you can easily append the cropyear but for the
designation you must remove the cropyear
William: change the derivative label in the top to designation
William: the name type should inherit to the new creation method
Warren: how will i know the name type...
Ching: that is in the wiki
Guoyou: why do you need the name type
William: We need it when creating new records
Guoyou: Basically, you just need the same name type for the generation you are creating
Warren: it is also useful in filtering search
Guoyou: my point is that it is just be one name type for all the lines
William: set a default, lets say F3, there should already be default already in your form (referring to germplasm, names, attributes tab). should get it from the ini file.
Warren: how about attributes?
Grace: sometimes you need to put different attributes. so just like the suggestion to put new column for name types, you should also do it for the attributes
Warren: Input Germplasm is also the form you could use for that but if you want another form for editing it, we can do so
Guoyou: my suggestion is you don't need to put the name and attributes in that form. if breeders really need to edit, do it while manipulating the list (sheet). add a column that says, short name etc
William: but some other breeder may want that
Warren: what william is suggesting is that we can select it in these tabs and the values with be automatically written in the sheet
Guoyou: that makes things worse. it's like you are forcing them to use the tabs.
William: but this can be customized. i mean if users do not want it, they can turn off the feature in the ini file.
Guoyou: but that is exactly what i do not want. what i want the is only the most commont tools will be added in those interface.
Guoyou: but you must not teach the users cause i doubt if anyone will request that. we should include only the good practices, not the bad ones
William: this has to be in the manual
Warren: Aien is working in the manuals
Grace: for the genebank, we need setting for the mgid, designation. but if we need to edit, we can do it in other interface, right? can you open a new source list and append it in the existing list?
ICIS Team: Yes.

coffee break

FIELDBOOK

Guoyou: can you create a list instead of opening a list?

Warren: yes
William: the menu should be new list -> type of list.
Guoyou: checklist should be part of the uploading function. remove the checklist icon. just add it as part of the loading process. you do not need to use it.
Warren: i use it to search germplasm
Aien: we can use the search germplasm icon. it's either we remove the checklist icon or the search germplasm icon as not the functions redundant.
Guoyou: but this is generating fieldbook.
Aien: warren is no yet using the fieldbook function. he is creating new list that will be used in generating fieldbook.
Grace: why don't we list now a menu that includes all the step for generating the fieldbook
Warren: ok we will just customize menu
Grace: what we do is we just retrieve an existing list then create fieldbook from it.
Aien: icis lite can already do that. warren can also show you how to do that. he is now showing how to create a list first before creating fieldbook because guoyou asked him to do so.
Warren: guoyou is just rearranging the order
Guoyou: my only point is that fieldbook should also work as standalone application in such a way that fieldbook will work even without anything else
William: make use of the trait group for the variable setup form
Guoyou: there shoud be agromy trait group then if you choose that group, it will show those traits that are only for agronomy
Warren: We can make it like a tree
Guoyou: give a user a shorcut.
Dec: that same function should also be present in the parent selection
William: First option is to create a fieldbook template, Initially it will contain the six columns from setgen. Then add your variates Then import your entries
Guoyou: You are not creating a template but you're creating a fieldbook, a template is something you use that already exists.
Grace:
-> Create a Fieldbook
1) You will be prompted with 2 options
A)Retrieve List in ICIS (already in ICIS)
B) Retrieve Germplasms from a Raw Excel File
2) Identify the Variables
A) Add Manually
B) Retrieve Variables from a DMS Study
C) Use a template file
3) Randomization
A) No Randomization at all
B) Replications
i) RCB
ii) Alpha Lattice
iii) Others
4) Ask whether to Generate Barcodes
A) Yes - Generate Barcodes then go to step 5
B) NO - go to step 5
5) Export to other type of files or Print

6) Save as Template

Guoyou: breeders need to have a smart user interface, a GUI for that.

lunch break

INVENTORY SYSTEM

Grace: Should also show other information about the requestor not just the name

Day 3: October 15, 2010

Continuation of Inventory System

Guoyou: after uploading the list pop up if you want to check the inventory
Grace: in my case, it's a matter of decision making. i will have to check the inventory first before using the fieldbook
Guoyou: i want the inventory system to be simple. just to check the availability of the seeds. inventory system should be separated.
Aien: grace just mentioned, that sometimes the administrator may not want to give away the seeds they have. so do they need to approve/dissaproved first after they send the application?
Grace: i need to know who are your target users so i can evaluate properly your application
Guoyou: breeders
Warren: we are talking to create here a separation of workflow not of the applications
Grace: the number of seed and the weight are the factors
Aien: aside from the institute, what information do you need?
Grace: person institute address private/public. all the information in the institute table should be filled up criteria for unavailability:
Grace: in genebank, there's a standard amount that is given, they dont need to specify the amount of seeds to be given. we status of the requestor's request. if your request has been opened, being process already etc

set a rule that can be the default for this instead of looking for all the seedlots.

ADMIN
receive notification/email of request
   |
open the transaction to be processed
   |
   approve/dissaprove request
   |
   view summary
   |
   send email notification

REQUESTOR
create a germplasm list/edit
   |
file a new seed transaction
   |
enter desired amount & scale
   |
   file seed request
   |
notification of availability/unavailability
   |
criteria of unavailability
1. below min balance
2. mls status
if available send email to administrator

Continuation of Parent Selection

Guoyou: remove third column, make first column tree view (if third column display result)
Warren: how about the methods, does that really matter?
William: to some, it does. you can use the method as a filter

PARENT SELECTION [for finding the right germplasm that you would like to advanced]
   |
specify study categories (optional to reduce the size) OR create germplasm list
   |
specify desired traits
   |
choose desired studies
   |
specify scale and method (optional)
   |
   show summary
   |
option to add other variables/traits in the output

SETGEN

Guoyou: setgen name isn't very informative
Grace: Generation No. need to have some notation that will show the number of generation
Ching: use G for generation instead of F
Vic: show list where this germplasm list-> display the folder names/display the path
Grace: sometimes you need to know from where nursery it is taken.
Grace: incoming seeds, how to know that they come in one set
Guoyou: we need to think of another field that will tell that they come from one set
Grace: should be easy seed health unit that "ah this is from 2008..." so they can search for the documents

ICIS Lite Interface Design Workshop