Mis-information

Options
Hi have signed up today and already have noticed that this is very inaccurate. Salmon is higher in fat than bacon according to this app... Not impressed

Replies

  • georgebrackley361
    georgebrackley361 Posts: 2 Member
    Options
    Worth mentioning that I checked this with a registered dietician, nutritionist, biomedical scientist and google. All of which disagree with the information provided.
  • VegjoyP
    VegjoyP Posts: 2,721 Member
    Options
    I go through all the choices and look for verified items. I also go on several web sites off MFP to check. Often times there is a margin of error even in the most ideal situation based on different aspects- volume vs weight, hydration, manufacturer, etc. Use discression and mindfulness. If your eating is relatively consistant you will also know by results. It is not an exact science, using this app has helped a mass amount of people.
    I f you do want the most precise and accurate information you may want to consider referring to several sources. You could even do the math for each macro to see if it adds up correct. Some nutrition labels do not include fiber calories for thier total caloric count.
  • WehttamThims
    WehttamThims Posts: 57 Member
    Options
    oh yeah, the database is an absolute sh*tshow. Who thought allowing anyone to enter anything was a good idea was probably the single worst decision the engineers ever came up with.

    You then have the "verified" ones, they have a green check mark next to them. That supposedly means the information is more correct than the other. We then have 50 "verified" sweet potato entries most of which say sweet potato's have 70 mg of sodium in them.
  • harper16
    harper16 Posts: 2,564 Member
    Options
    oh yeah, the database is an absolute sh*tshow. Who thought allowing anyone to enter anything was a good idea was probably the single worst decision the engineers ever came up with.

    You then have the "verified" ones, they have a green check mark next to them. That supposedly means the information is more correct than the other. We then have 50 "verified" sweet potato entries most of which say sweet potato's have 70 mg of sodium in them.

    How would you have set up the database?
  • lemurcat2
    lemurcat2 Posts: 7,885 Member
    Options
    harper16 wrote: »
    oh yeah, the database is an absolute sh*tshow. Who thought allowing anyone to enter anything was a good idea was probably the single worst decision the engineers ever came up with.

    You then have the "verified" ones, they have a green check mark next to them. That supposedly means the information is more correct than the other. We then have 50 "verified" sweet potato entries most of which say sweet potato's have 70 mg of sodium in them.

    How would you have set up the database?

    Actually, one superior way would be how it used to be, with the USDA entries easy to identify (those were added by MFP). Having the rest available is better when logging packaged items, of course, and it's easy to compare to the package.

    In a perfect world (IMO), they wouldn't allow people to make their homemade meals public, as there's a surprising lack of common sense that "one plate" of someone else's "homemade chicken and potatoes" might not have the same cals or macros as anyone else's. But eh, although those clutter up the database it's not very hard to learn to navigate, IMO.
  • bpetrosky
    bpetrosky Posts: 3,911 Member
    edited June 2020
    Options
    lemurcat2 wrote: »
    harper16 wrote: »
    oh yeah, the database is an absolute sh*tshow. Who thought allowing anyone to enter anything was a good idea was probably the single worst decision the engineers ever came up with.

    You then have the "verified" ones, they have a green check mark next to them. That supposedly means the information is more correct than the other. We then have 50 "verified" sweet potato entries most of which say sweet potato's have 70 mg of sodium in them.

    How would you have set up the database?

    Actually, one superior way would be how it used to be, with the USDA entries easy to identify (those were added by MFP). Having the rest available is better when logging packaged items, of course, and it's easy to compare to the package.

    In a perfect world (IMO), they wouldn't allow people to make their homemade meals public, as there's a surprising lack of common sense that "one plate" of someone else's "homemade chicken and potatoes" might not have the same cals or macros as anyone else's. But eh, although those clutter up the database it's not very hard to learn to navigate, IMO.

    The 'cardinal sin' of MFP's crowdsourcing plan was that they made it too permissive initially, and never implemented an effective plan for curating the database and culling bad entries. Wikipedia, probably the best Web 2.0 era crowdsourcing system, went through a pretty similar phase years ago, and they only solved it with a very strong volunteer curation team that tries to keep the garbage out as much as they can.

    I believe that initially MFP user entries were public by default, and they changed it several years ago, but I don't know when that would have happened.

    Better search and filter options would be helpful (say options to filter for USDA entries, product country, etc) but that presumes that the data was captured on entry.

    Someday, MFP may have to rebuild their database model and start over applying some best practices they learned over the years to support a more sustainable design. They could keep the present database as a 'legacy' source until the new one gets populated enough. But that could confuse users as much as help them. I doubt UA is that invested in improving MFP though.