start values do not take sign of fixed parameters into account

[NilsPetras]

To avoid reversely keyed factors stemming from reversely keyed indicator items, I attempted to use "factor=~-1*Item1 + ..." style syntax for model identification (mock example):

My code:

HS.model <- ' visual  =~ -1*x1 + x2 + x3
              textual =~ x4 + x5 + x6
              speed   =~ x7 + x8 + x9 '
fit <- cfa(HS.model, data = HolzingerSwineford1939)

Standard code:

HS.model <- ' visual  =~ x1 + x2 + x3
              textual =~ x4 + x5 + x6
              speed   =~ x7 + x8 + x9 '
fit <- cfa(HS.model, data = HolzingerSwineford1939)

This led to convergence issues on my own data. Comparing to the standard model identification, the further start values assigned to the factor identified by -1* consistently have the wrong sign:

My code start values (lambda):
x1 -1.000 0.000 0.000
x2 0.778 0.000 0.000 <= inconsistent (unchanged) sign
x3 1.107 0.000 0.000 <= inconsistent (unchanged) sign
x4 0.000 1.000 0.000
x5 0.000 1.133 0.000
x6 0.000 0.924 0.000
x7 0.000 0.000 1.000
x8 0.000 0.000 1.225
x9 0.000 0.000 0.854

Standard code start values (lambda):
x1 1.000 0.000 0.000
x2 0.778 0.000 0.000
x3 1.107 0.000 0.000
x4 0.000 1.000 0.000
x5 0.000 1.133 0.000
x6 0.000 0.924 0.000
x7 0.000 0.000 1.000
x8 0.000 0.000 1.225
x9 0.000 0.000 0.854

I specified custom start values to solve this, which is inconvenient but works. I would love to see this dealt with by lavaan itself.